0 a
r
O A K RIDGE N A T I O N A L LABORATORY ,d!--{ 1
operated by M. .I, U N I O N CARBIDE CORPORATION
f o r the
U.S. ATOMIC ENERGY COMMISSION
ORNL-TM- 915
/8 I 3 445b 0548825 5 I
STUDIES IN HEAT TRANSFER AND FLUID MECHANICS
PROGRESS REPORT FOR PERIOD JANUARY 1 - SEPTEMBER 30, 1963
- H. W. Hoffman J. J. Keyes, Jr.
NOTICE T h i s document con ta ins i n fo rma t ion of a pre l iminary nature and was prepared p r i m a r i l y for i n te rna l use at the Oak Ridge Na t iona l Laboratory . It i s sub jec t t o r e v i s i o n or co r rec t i on and therefore does not represent a f i n a l report. i n fo rma t ion i s not t o be abstracted, rep r in ted or o the rw ise g i ven p u b l i c d i s - semina t ion wi thout t he approval o f the ORNL patent bronch, L e g a l and In for - ma t ion Contro l Department.
T h e -1
I LEGAL This report was prepared os on account of Government sponsored work.
nor the Commission, nor ony psrson acting on behalf of the Commission:
A. Makes any warranty or representation, expressed or implied, wi th respect t o the accurocy,
completeness, or usefulness of the information contoined i n th is report, or thot the use of
ony informotion, apparatus, method, or process disclosed in th is report may not infringe
privately owned rights; or
Assumes ony l iab i l i t ies wi th respect to the use of, or for domoges resulting from the use of
any informotion, opparotus, method, or process disclosed in th is report.
Neither the United States,
8.
As used in the above, "person acting on behalf of the Commission'' includes any employee or
controctor of the Commission, or employee of such contractor, to the extent thot such employee
or contractor of the Commission, or employee of such contractor prepores, disseminates, or
provides occess to, ony informotion pursuont to his employment or controct wi th the Commission,
or h i s employment wi th such controctor.
3
ORNL- TM- 915
Contract No. W-7405-eng-26
Reactor Division
. STUDIES' I N HEAT TRANSFER AND'FLUID MECHANICS PROGRESS REPORT FOR PERIOD JAN. 1- SEFT. 30, 1963
H. W . Hoff'man J. J. Keyes, Jr.
OCTOBER 1964
U.S. ATOMIC ENERGY COMMISSION ~ 3 4456 0548825 5
E
i
iii
CONTENTS
Page
S U M M A R Y . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . v
1. INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . 1
2. BOILING HEAT TRANSFER AND TWO-PHASE FLOW . . . . . . . . . . . . 5 2.1. Boiling Potassium Heat Transfer . . . . . . . . . . . . . . 5
2.1.1. Forced-Convection System . . . . . . . . . . . . . 5 2.1.2. Superheat Pnenomena . . . . . . . . . . . . . . . . 21
2.2. Water Boiling i n a Multirod Geometry . . . . . . . . . . . 27 /
2.3. General Boiling Studies . . . . . . . . . . . . . . . . . . 30 2.3.1. Predict ion of C r i t i c a l Heat Fluxes . . . . . . . . 30
2.3.2. Randomness i n the C r i t i c a l Heat Flux. . . . . . . . 32
2.3.3. In t e rna l Heat Transfer i n a Horizontal Tube
Immersed i n a Water Pool . . . . . . . . . . . . 34 2.4. Swirl-Flow Heat Transfer . . . . . . . . . . . . . . . . . 35
2.4.1. Heat-Transfer Charac te r i s t ics of Ethylene Glycol. . 35 2.4.2. Swirl-Flow Boiling with Net Vapor Generation . . . 40
3. FLOW DYNAMICS AND TURBULENCE . . . . . . . . . . . . . . . . . . 41 3.1. Vortex F lu id Mechanics . . . . . . . . . . . . . . . . . . 41
3.1.1. Gas Dynamics Studies . . . . . . . . . . . . . . . 42
3.1.2. Magnetohydrodynamic Studies . . . . . . . . . . . 60 3.2. Turbulence Transport Studies > . . . . . . . . . . . . . . . 78 3.3. Boundary-Layer Transient Phenomena . . . . . . . . . . . . 86
4. THERMOPHYSICAL PROPERTIXS . . . . . . . . . . . . . . . . . . . . 97 4.1. Alka l i Liquid Metals . . . . . . . . . . . . . . . . . . . 97
4.1.1. Thermal Conductivity of Lithium . . . . . . . . . . 97 4.1.2. Surface Tension of Potassium . . . . . . . . . . . 101
4.1.3. Contact Angle Determinations . . . . . . . . . . . lo5 4.2. Miscellaneous Materials . . . . . . . . . . . . . . . . . . lo7
4.2.1. E l e c t r i c a l R e s i s t i v i t y of Brass . . . . . . . . . . lo7 5 . BIBLIOGRAPHY. . . . . . . . . . . . . . . . . . . . . . . . . . . 109
i
,
. 7
\
,
v
I SUMMARY
Boiling Heat Transfer and Two-Phase Flow
Boiling Potassium Heat Transfer. An experimental study of forced-
convection sa tura ted boi l ing with potassium w a s undertaken t o deternine
c r i t i c a l heat f lux l i m l t s , heat- t ransfer coef f ic ien ts , and two-phase
pressure drop c h a r a c t e r i s t i c s . Data have been obtained f o r boi l ing in-
s ide of two v e r t i c a l tubes of d i f f e r e n t diameters (0.87-in. and 0.325-in.);
the heat f lux c a p a b i l i t y of these b o i l e r s w a s 40,000 and 5OO,OOO Btu/hr-ft:
respectively.
I n the lower-flux ( la rger diameter) t e s t sect ion, tube-wall tempera-
t u r e o s c i l l a t i o n s of very high amplitude ( + l 5 O t o +200"F) were observed
during some periods of operation; these o s c i l l a t i o n s were regular i n f r e -
quency and pers i s ted over long t imes. temperature f luc tua t ion) w a s found t o be possible only when vapor bubbles
ex is ted a t the b o i l e r i n l e t .
Stable operation (minimal w a l l -
C r i t i c a l heat-flux values ranged from -32,500 Btu /hr - f t2 a t an i n l e t
mass flow of 0.15 x IO5 l b / h r . f t 2 t o -345,000 Btu /hr - f t2 a t G = 1 . 3 X lo5
lb /hr . f t2 . The r e s u l t s agreed well with the cor re la t ion of Lowdermilk, \ Lanzo, and Siege1 based on data f o r the forced-convection saturated b o i l -
ing of water.
the other a l k a l i l i q u i d metals. When s t a b l e high-temperature o s c i l l a t i o n s
occurred, the c r i t i c a l heat f lux w a s reduced by about 40%.
It is not expected t h a t t h i s comyarison w i i l be v a l i d f o r
Heat-transfer data f o r f lux l e v e l s below the c r i t i c a i show a l e s s e r
dependence of (q/A) on LSL' than has been found with water. Further, the
-data obtained i n the low-flux b o i l e r f a l l s i g n i f i c a n t l y below those ob-
ta ined with the h i g h - f l m b o i l e r . The l a t t e r r e s u l t s agree reasonably
with the preliminary data of BrooJks (General Elec t r ic Company).
Pressure-loss da ta (as the r a t i o of t o t a l two-phase t o equivalent
l iquid-only a t the same i n l e t mass flow) have been cor re la ted using the
Lockhart-Martinelli parameter. A consis tent v a r i a t i o n was observed such
t h a t the pressure-loss r a t i o increased with increasing e x i t qual i ty . E x i t
q u a l i t i e s ragged between 10 and 86$.
v i
Superheating i n a l k a l i l i q u i d metals a t t h e i r normal bo i l ing tem- peratures w a s ca lcu la ted t o vary from 2 5 8 O ~ f o r sodium down t o 6 7 " ~ f o r
cesium i n the presence of c a v i t i e s which would allow only a 30°F super-
heat i n water a t atmospheric pressure . It w a s a l s o shown t h a t t h e super-
heating i n potassium might be g rea t e r than indicated by t h e ca lcu la t ion ,
s ince potassium is general ly c leaner , wets t h e metal b e t t e r , and more
r ead i ly f i l l s surface c a v i t i e s than does water.
Vapor present a t t h e b o i l e r i n l e t may produce s t ab le operat ion by
el iminat ing the high superheat necessary f o r bubble i n i t i a t i o n . On the
other hand, considerat ions i n terms of an annular flow l ead t o t h e obser-
va t ion t h a t evaporation from a t h i n , superheated l i q u i d f i l m on the w a l l
t o t h e c e n t r a l vapor flow would be s u f f i c i e n t t o account f o r t he observed
heat t r a n s f e r .
Water Boiling i n a Multirod Geometry. Apparatus has been assembled
f o r an experimental study i n t o the nagnitude of t h e c r i t i c a l hea t f l ux
and the two-phase pressure loss f o r the forced-convection bo i l ing of water
a t atmospheric pressure flowing along t h e axis of a heat 7-rod c l u s t e r .
The e f f e c t s of rod spacing, i n l e t subcooling, and Gutlet q u a l i t y w i l l be
examined. It i s hoped t o resolve some discrepancies i n the l i t e r a t u r e
as t o the loca t ion of t h e burnout region, as wel l as t o provide da ta of
value t o reac tor design.
General Boiling Studies . An addi t ive -type c o r r e l a t i o n (combining
pool-boiling w i t h forced-convectlon e f f e c t s ) i s described f o r use i n pre-
d i c t ing the c r i t i c a l heat f l ux f o r forced-convection bo i l ing .
found t h a t 9% of t he predicted values agreed w i t h ava i l ab le experimental
data t o within 49; ' t h i s is pointed out as providing a b e t t e r representa-
t i o n of ex i s t ing experimental data than given by previously ava i lab le
co r re l a t ions .
It w a s
An experimental study i n t o t h e inherent randomness i n t h e c r i t i c a l
heat flux was conducted using an apparatus cons is t ing of a hor izonta l ,
e l e c t r i c a l l y heated tube i n a pool of s a tu ra t ed water. It w a s found t h a t
t he re ex i s t ed an uncer ta in ty i n the c r i t i c a l heat f l ux value which approxi-
mated t h a t predicted by Zuber's hydrodynamic i n s t a b i l i t y theory and that
v i i
t h e nature and condition of the b o i l e r surface can cons t i tu te an impor-
t a n t influence on the c r i t i c a l heat f lux .
An experiment with boi l ing on the inside only of a horizontal tube
submerged i n a l i q u i d pool w a s completed.
were open, there resu l ted i n most t e s t s a pulsat ing flow i n which l i q u i d
flow inward toward the tube center a l t e r n a t e d w i t h an outward expulsion
of liquid-vapor mixture.
Since the ends of t h e tube
This w a s found t o occur pr imari ly w i t h tubes
of large L/d r a t i o ; f o r tubes of small L/d r a t i o , behavior similar t o
that observed i n other geometric s i t u a t i o n s i n a t t a i n i n g the c r i t i c a l
f lux w a s observed.
Swirl-Flow Heat Transfer. Swirl-flow heat t r a n s f e r with ethylene
glycol has been s tudied as p a r t of a general invest igat ion o f the i n f l u -
ence of coolant thermophysical propert ies on heat t r a n s f e r w i t h t h i s mode
of flow. E l e c t r i c a l l y heated, horizontal tubes containing twisted-tape
s w i r l generators varying i n t w i s t (tube diameters per 180-deg t u r n ) from
2.2 t o 12.1 were used. Over the experimental ranges of t h i s study, the
c r i t i c a l heat f luxes f o r saturated pool boi l ing, axial-flow l o c a l b o i l -
ing, and swirl-flow l o c a l boi l ing were found t o be d / 3 , 1/2, and 3/4 of
those f o r water a t t h e same conditions of geometry, velocity, pressure,
and bulk temperature. The predict ions of the addi t ive cor re la t ions were
i n good agreement w i t h these data . A cor re la t ion describing nonboiling
swirl-flow data for both ethylene glycol and water t o within +12$ w a s
developed.
high v e l o c i t i e s and heat f luxes were higher than predicted by ex is t ing standard cor re la t ions .
Data on nonboiling axial-flow hea t - t ransfer coef f ic ien ts a t
An apparatus i s being assembled f o r two-phase swirl-flow boi l ing
s tudies .
Flow Dynamics and Turbulence
Vortex Fluid Mechanics. I n considering t h e use of vortex flow f o r
such advanced energy conversion appl icat ions as the gaseous core nuclear
reactor and t h e magnetohydrodynamic power generator, knowledge of the
in te r re la t ionship of the vortex s t rength and the turbulent energy d i s s i -
pat ion is of fundamental importance. Experimental gas dynamics s tud ies
v i i i
have revealed, f o r example, t h a t a high l e v e l of tu rbulen t shear i n the
concave w a l l boundary layer se r ious ly limits the vortex s t r eng th a t t a i n -
able with an energy expenditure which i s deemed reasonable from t h e stand-
point of t he appl ica t ions . Furthemore, turbulence which may e x i s t i n
the i n t e r i o r of t he flow f i e l d i s a l s o undesirable from t h e standpoint of
separat ion e f f ic iency . Spec i f ica l ly , these s tud ie s have de l inea ted the
e f f e c t of tube diameter, mass flow r a t e , i n j ec t ion ve loc i ty .and geometry,
and p res swe on the vortex s t rength , as w e l l as the e f f e c t of turbulence
on the separat ion c h a r a c t e r i s t i c s of vortex flow. Techniques examined
f o r possible reduction i n boundary l aye r turbulence include uniform w a l l
bleed-off and s l i t bleed-off . Neither technique,. however, produced a
s ign i f i can t e f f e c t on the vortex s t rength . The vortex s t rength was ob-
served t o increase s i g n i f i c a n t l y w i t h extreme w a l l cool ing. Experiments
i n which vortexes were generated by flow through a poroas w a l l , t h e pores
of which were or ien ted near ly t angen t i a l ly , were encouraging, but ind i -
ca ted t h a t an extremely la rge number of very small, c lose ly spaced pores
would l i k e l y be necessary t o a t t a i n tne des i red l imi t ing condi t ion of a "rotat ing" w a l l of gas j e t s .
A n ana iy t i ca l and experimental study concerned with t h e possible
appl ica t ion of a hydromagnetic s t a b i l i z a t i o n technique f o r vortex f l o w
w a s i n i t i a t e d . Analysis based on a nondissipative model revealed t h a t
magnetic s tab i l iza+, ion i s possible , using an a x i a l magnetic f ie ld , w i t h -
out a l t e r i n g the mean flow. Analysis of a more r e a l i s t i c case, dissi-
pat ive vortex flow, w a s i n f t i a t e d . AE exploratory experimental i nves t i -
ga t ion i n t o je t -dr iven, confined -mrtex-type flow, using an e l e c t r o l y t i c
so lu t ion (concentrated NH,C1) as t h e working f l u i d and a 62-kilogauss
axial magnetic f i e l d , demonstrated t h e s i g n i f i c a n t s t a b i l i z i n g influence
of t h e magnetic i n t e rac t ion . For example, t h e per iphera l t angen t i a l
Reynolds modulus a t t r a n s i t i o n t,o i n s t a 5 i l i t y on t h e concave w a l l was .
increased by the magnetic f i e l d , as was t h e recovery of in j ec t ion veloc-
i t y as t angen t i a l ve loc i ty . An add i t ioca l magnetic e f f e c t of i n t e r e s t
is' t he s t a b i l i z a t i o n of t he boundary l aye r on the c i r c u l a r end w a l l , and
t h e r e su l t i ng increase i n t h e r a t i o of t angen t i a l t o r a d i a l ve loc i ty .
ix
Turbulent Transport Studies. A n experimental approach t o the pro-
. blem of Reynolds s t r e s s determination i s proposed based on an improved
theory r e l a t i n g the s i g n a l of a hot-wire anemometer t o the Reynolds
s t r e s s e s .
momentum conductivity of the f l u i d . ]
analog computer has been developed and ca l ibra ted , and s tab le hot wire
probes have been fabr ica ted .
t e s t e d which meets s t r ingent f l u i d cleanl iness , deaeration, and thermal
s t a b i l i t y requirements. Preliminary r e s u l t s f o r pipe flow of water a t
Reynolds modulus of 158,000 have indicated t h a t the l e v e l of turbulence
i n water flows i s possibly l e s s than an t ic ipa ted from exis t ing measure- ments i n a i r .
[The Reynolds s t r e s s e s represent the eddy contr ibut ion t o the
A su i tab le instrument employing an
A water-flow loop has been constructed and
Boundary-Layer Transient Phenomena. A n invest igat ion i n t o the nature
of turbulent i n s t a b i l i t i e s i n boundary layers i s i n progress with the aim
of del ineat ing t h e mechanism of surface thermal t r a n s i e n t generation as
a p o t e n t i a l source of fatigue-type f a i l u r e and/or accelerated corrosion
i n reactor heat exchange systems. Exploratory experiments employing an
e l e c t r i c a l l y heated 2- in . pipe as the t e s t sec t ion with deaerated water
i n turbulent flow have been c a r r i e d out f o r the purpose of instrumentation
development and evaluation. These experiments have demonstrated t h a t "gun-
bar re l" type surface thermocouples w i l l provide quant i ta t ive measurements
of the surface thermal t r a n s i e n t s and t h a t the integrated read-out system
[employing amplification, recording (magnetic tape and oscil lographic ) , and spec t ra l a n a l y s i s ] i s s a t i s f a c t o r y .
d i r e c t determination of instantaneous hea t - t ransfer r a t e s a t a f l u i d - s o l i d
interface has been developed. A simple mathematical model which takes
i n t o account the breaking up of the viscous sublayer and eddy penetrat ion
has been compared with the preliminary experimental r e s u l t s .
A hot film surface probe for
Thermophys i c a l Properties
Alkali Liquid Metals. The measurement of the thermal conductivity
of molten l i thium w a s completed; f i n a l r e s u l t s showed a l i n e a r increase
with temperature from 25.6 Btu/hr. f t O F a t 6000~ t o 35.0 Btu/hr - f t O F a t
D
X
1500°F. lower temperature t o 216 a t the upper end of the range.
with sodium and potassium, the thermal conduct ivi ty of l i th ium showed a
pos i t ive temperature c h a r a c t e r i s t i c . A comparison of these r e s u l t s with
values derived from recent e l e c t r i c a l r e s i s t i v i t y and thermal d i f f u s i v i t y
measurements indicated a similar t r end with temperature, although i n mag- nitude t h e cur ren t da ta f e l i 4 t o lo'$ below the data of t h e o ther i nves t i -
ga tors .
A de t a i l ed e r r o r ana lys i s indicated an accuracy of 57% a t the
I n cont ras t
Preliminary valuces were obtained f o r t he surface tens ion of potassium
against helium over the temperature span 70 t o 308"c using a maximum bub-
b l e pressure technique. The r e s u l t s were s i g n i f i c a n t l y higher than da ta
obtained by other inves t iga tors using c a p i l l a r y r i s e techniques.
Experiments were i n i t i a t e d on measuring the contact angle f o r potas-
sium drople ts aga ins t various metals i n order t o determine the extent t o
which surfaces a re wetted by potassium. I n preliminary s tud ie s on a pol-
ished type 316 s t a i n l e s s s t e e l surface, it w a s observed t h a t t h e angle of
contact var ied from 120-deg a t 70°C t o 0-deg at 500°C some 95 minutes
l a t e r .
contact angle of 50-deg .
A second drople t placed on the surface a t 180"~ had an i n i t i a l
Miscellaneous Mater ia ls . The e l e c t r i c a l r e s i s t i v i t y of seamless
70-30 brass tubing w a s experimentally determined and found t o vary from
7:4 pohm-ern a t 100°C t o 1 ~ 8 pohm-cm a t 500°C.
i
C
a . ... ..
1
STUDIES I N HEAT TRANSFER
PROGRESS REPORT FOR PERIOD
H . W . Hoff'man
AND FLUID MECHANICS JAN.. 1 - SEE". 30, 1963
J. J . Keyes, Jr.
1. INTRODUCTION
The design of any spec i f i c nuclear reac tor r e s t s on such knowledge
as is ava i lab le at the time coupled with t e s t data e a s i l y obtainable
during the design per iod. Detailed design commitments cannot be founded
on projected t e s t programs of unknown durat ion and uncertain outcome.
However, i n accomplishing h i s function, t h e design engineer i s o f t en
placed i n the un fa i r pos i t i on of having t o ex t rapola te beyond the cur-
ren t s ta te -of -knowledge, hoping t h a t h i s "guesses " a re educated and
l o g i c a l and t h a t h i s decisions w i l l be j u s t i f i e d by successful operation
and/or l a t e r corroborative developments.
of a s s i s t i n g a rap id r a t e of advance i n successive reac tor designs i s
t o an t i c ipa t e design problems by conducting s tudies , independent of any
spec i f ic reac tor , on problems common t o general reac tor types . One such
important a r ea of problems i s t h a t of Heat Transfer and Flu id Mechanics.
This includes extending experimental inves t iga t ions outs ide the range of
i m e d i a t e i n t e r e s t s o as t o generate broader empirical cor re la t ions , as
wel l as the development of t h e o r e t i c a l and experimental knowledge f o r new geometries, coolants , and thermal and hydrodynamic condi t ions.
It follows then t h a t one way
In addi t ion t o t h i s pos i t ive e f f o r t towarding speeding advances i n
reac tor design, the inves t iga tor must be a l e r t f o r and must pursue par-
t i a l l y hidden ind ica t ions of unsuspected hazards and d i f f i c u l t i e s which
might otherwise be overlooked i n the s t r e s s and rush accompanying the
design and development of a reac tor system. The i d e n t i f i c a t i o n of t he
p o s s i b i l i t y of se r ious s t r e s s cracking and/or enhanced corrosion i n heat
exchanger tubes operated a t unusually high heat f l ux with turbulen t
cooling o r heating by a f l u i d of moderate t o high thermal conduct ivi ty
i s one example of t h i s type of cont r ibu t ion .
I n recognition of these f a c t s , t he Engineering Development Branch
of t he U. S. Atomic Energy Commission has supported f o r the pas t severa l
2 a
years a program i n heat t r a n s f e r and f l u i d mechanics a t the Oak Ridge
National Laboratory aimed a t point ing t h e way toward advances i n design
of both new and e x i s t i n g reac tor concepts.
t he areas of turbulence, two-phase flow, and bo i l ing heat t r a n s f e r .
Th i s work i s concentrated i n
Turbulence s tudies include t h e measurement of Reynolds s t r e s s e s and
t h e development of techniques f o r making such measurements, t h e e f f e c t s
of hydrodynamic con t ro l and magnetic f i e l d s i n s t a b i l i z i n g a high-strength
vortex flow, and the influence of tu rbulen t f l uc tua t ions i n inducing
thermal t r ans i en t s a t a heat - t r ans fe r r ing surface. While fundamental i n
form, these invest igat ions a r e or ien ted t o usefu l and r e a l i s t i c geometries
i n p r a c t i c a l s i t u a t i o n s . Thus, as mentioned previously, t he last-named
area has included t h e determination of l o c a l fa t igue cracking and corro-
s ion e f f e c t s on a metal w a l l subjected t o temperature o s c i l l a t i o n s and
w i l l be extended t o r e l a t e t he influence of upstream flow d i s tu rbe r s
(elbows, o r i f i c e s , e t c . ) i n cont r ibu t ing t o these o s c i l l a t i o n s .
A s ign i f i can t e f f o r t is d i rec ted t o t h e s tudy of bo i l ing phenomena
with water and potassium (pr imari ly) f o r a v a r i e t y of flow and geometric
condi t ions. Included are the determination of c r i t i c a l heat f l u x and
hea t - t ransfer coe f f i c i en t s w i t h potassium, t h e e f f e c t s of s w i r l flow on
burnout under subcooled and net vapor generation condi t ions, t h e influence
of rod spacing on the c r i t i c a l heat f l u with bo i l ing along a rod bundle,
and two-phase pressure l o s s i n channels w i t h forced-convection sa tu ra t ion
boi l ing of potassium.
c i p a l among these being an inves t iga t ion of l i q u i d superheat w i t h t he
a lkal i l i q u i d metals as a source of i n s t a b i l i t y i n the bo i l ing process.
Experiments and t h e o r e t i c a l s tud ie s of t he pressure drop and regime t r a n -
s i t i o n s i n the two-phase flow of l i q c i d metals p a r a l l e l t o rod b h d l e s
and of l iquid-metal vapor condensation a r e planned.
A number of r e l a t e d e f f o r t s have developed, p r in -
I n support of a l l of t h i s work is the experimental determination of
t he thermophysical p roper t ies of coolants and containment materials. A t
present these s tud ies a re r e s t r i c t e d ' t o t he measurement of t h e surface
tension, densi ty , and thermal conduct ivi ty of t he alkali l i q u i d metals
a t temperatures t o 2000°F.
.
3 I
This report i s the f irst i n a s e r i e s providing a periodic summation
of the r e s u l t s obtained i n these s tud ies . The second report w i l l cover the period through June 30, 1964; t h e r e a f t e r , these w i l l be issued semi-
annually.
issued since the inception of t h i s program i s t o be found i n the b i b l i o -
A complete l i s t i n g of reports and open-l i terature publicat ions
graphy.
a
.
i
5
2. BOILING HEAT TRANSFER AND TWO-PHASE FLOW
2.1. Boiling Potassium Heat Transfer
.
2.1.1. Forced Convection System
A. I. Krakoviak H . W . Hoffman
Introduction
An understanding of t he thermal and f l u i d mechanical c h a r a c t e r i s t i c s
of bo i l ing a l k a l i metals i s cf grea t importance i n designing l ight-weight
Rankine-cycle nuclear-energy systems t o s a t i s f y a u x i l i a r y power require-
ments i n s a t e l l i t e o r o ther space vehic les . Knowledge of burnout heat
flux l i m i t s , hea t - t ransfer coe f f i c i en t s , two-phase pressure losses , and
flow s t a b i l i t y c r i t e r i a i s of immediate i n t e r e s t ; and the experimental
program discussed below i s intended t o provide such information with bo i l -
ing potassium.
Attempts a t pred ic t ion of the c r i t i c a l heat f l ux with the l i q u i d
a l k a l i metals i n forced-convection bo i l ing a t sa tu ra t ion conditions, '
using cor re la t ions developed f o r water and organic f l u i d s , have yielded
widely disparate r e s u l t s . Thus, from the generalized expressions of
Griff i th , ' the estimated peak heat f l u x (at 50% e x i t qua l i ty ) i s -2 X lo8
Btu/hr.ft2; while an addi t ive procedure recent ly developed by Ga~nb i l l ' , ~
suggests a value of -10 x lo6 Btu/hr . f t2 .
values f o r t he c r i t i c a l heat f lux were not ava i lab le a t the inception of
t h i s program.
Experimentally determined
'W. R . G a m b i l l and H . W . Hoffman, "Boiling Liquid-Metal Heat Trans-
'P. G r i f f i t h , "The Correlat ion of Nucleate Boiling Burnout Data, ' I
%. R . Gambill, "Generalized Predic t ion of Burnout Heat Flux f o r Flow-
f e r , " American Rocket Society Paper No. 1737-61, May 1961.
ASME Preprint No. 57-HT-21; see a l so MIT Tech. Report No. 9, March 1957.
. ing, Subcooled, Wetting Liquids, " Chemical Engineering Progress Symposium Series , 59(41): 71-87 (March 1963).
..
6
Apparatus
The experimental system assembled f o r t h e measurement of t h e c r i t i c a l
heat f l ux with potassium is depicted schematically i n Fig. 2.1. Liquid
potassium was pimped from a cooled reserv ior through a v e r t i c a l l y mounted
b o i l e r t o a l iquid-vapor separator .
provided the pressure drop necessary t o e l iminate hydrodynamic coupling
between the b o i l e r and the pumping system. The flow r a t e s of t h e l i q u i d
and t h e vapor a f t e r condensation were determined both volumetr ical ly i n
ca l ib ra t ed hold tanks and dynamically by electromagnetic flow meters; the
b o i l e r e x i t qua l i t y w a s based on these measurements. A t angen t i a l i n l e t
t o t he separator caused the necessary. cen t r i fuga l force for separat ion of
the en ter ing vapor- l iqcid mixtures; a t low flows, g rav i ty separat ion ap-
peared e f f e c t i v e .
condenser. Oxygen w a s kept at a low concentration by the continuous by-
pass c i r cu la t ion of a small stream of l i q u i d potassium through a heated
t i tanium sponge t r a p .
A r e s t r i c t i o n a t t h e b o i l e r i n l e t
A finned c o i l cooled by an a i r b l a s t served as the
Boilers of two designs have been used thus far; these are shown i n
I n i t i a l measurements* were made with a cut-away schematic i n Fig. 2.2.
t e s t sec t ion constructed of a 6-f t long, 1-in.-OD (0.87-in.-ID) type 347 s t a i n l e s s s t e e l tube . The lef t -hand por t ion of Fig. 2.2 depic t s a typ ica l
u n i t length of t h i s b o i l e r . Heat w a s suppl ied by a s e t cf six 3-in.-ID X
12-in. -long clamshell hea te rs centersd on t h e b o i l e r tube . Thermocouples
(8-mi1-d.i,m Ft-pi; 10% R h ) were welded t o t h e outs ide tube surface a t t h e
approximate a x i a l mid-point of each hea ter ; thermowslls and pressure t aps
f o r es tab l i sh ing the f lu id condi t ions were loca ted a t t h e t e s t s ec t ion
entrance and e x i t .
In . - I D sec t ion ; and a t the o u t l e t end, t o a 0.625-in. - I D l i n e .
mum heat f l u a t t a inab le with t h i s configurat ion w a s -40,000 Btu/hr. f t 2 .
A t t he i n l e t , t h e b o i l e r tube w a s welded t o a 0.245-
The maxi-
The second b o i l e r is shown as the right-hand por t ion of F ig . 2.2.
This was a 44-in.-long, 0.325-in.-ID (0.028-in.-wall th ickness) type 347
%. W . Hoff'man and A. I. Krakoviak, Forced-Convection Saturat ion Boil- i cg of Potassium a t Near-Atmospheric Pressure, pp. 182-203, "Proceedings of 1962 High-Temperature Liquid-Metal Heat Transfer Technology Meeting," USAM: Report Ba-756, Brookhaven National La,boratory.
7
UNCLASSIFIED ORNL-LR-DWG 46124A
VACUUM
PRESSURE
L - VAPOR '
I MIXING CHAMBER
-VENT CONDENSER
CONDENSER 0 MIXING
CHAMBER
I I - l l U ' b " TUBE
HOLD
L E V E L INDICATOR
4- A i
FLOWMETER
AIR I v LIQUID
RESERVOIR I 1 1 'PUMP'
FREEZE VA LV E
I I SUMP TANK
HOT TRAP
Fig. 2.1. Loop Schematic: Forced-Convection Boiling with Potassium.
8
UNCLASSIFIED ORNL-LR-DWG. 761 13R
Fig. 2.2. Test Sections f o r Boiling Potassium m ' e r i m e n t .
.
. 9
s t a i n l e s s s t e e l tube brazed within a segmented copper sleeve of 5-in.
outside diameter. The twenty-one, 2-in. - thick OFHC copper disks com-
pr i s ing t h i s sleeve were separated by O.ll9-in. gaps t o minimize the
a x i a l heat flow. Each d isk w a s heated ex terna l ly by an individual ly con-
t r o l l e d clamshell heater ; the converging geometry of the heat-flow path
r e s u l t s i n s u b s t a n t i a l l y augmented heat f luxes a t the fluid-metal i n t e r -
face f o r seasonable heater temperatures. This b o i l e r design permitted
heat f luxes of t h e order of 500,000 Btu/hr.ft2.
gradient i n each copper block w a s measured with three logarithmically
spaced 0.040-in. -OD sheathed Chromel-Alumel thermocouples.
couple readings were extrapolated t o give the temperature a t the boi l ing
surface and, i n combination with the copper conductivity, the heat f luxes
through each copper segment. Final ly , the mass of the copper disks pro-
vided s u f f i c i e n t thermal capacity t o insure against physical burnout of
the b o i l e r tube.
The r a d i a l temperature
These thermo-
.
Ope r a t i n g Charac te r i s t ics
Outside tube-wall temperature t r a c e s obtained i n runs B6.3 through
B6.6 during approach t o "burnout" are shown i n Fig. 2.3; thermocouples 5 and 6 were located near the b o i l e r e x i t . A s indicated i n t h i s f igure,
\ the c r i t i c a l heat f lux w a s a t ta ined by progressively reducing the flow,
while maintaining a constant heat f l u x l e v e l i n the b o i l e r .
point w a s indicated by the automatic in te r rupt ion of b o i l e r power when the
w a l l temperature during any o s c i l l a t i o n exceeded a prese t value.
Ser ies B experiments, t h i s temperature l i m i t w a s 1750'F; i n some e a r l y
runs temperature excursions of as much as 350°F above the mean were noted. In the high-flux b o i l e r (Series A ) , an increase i n w a l l temperature of
50°F w a s taken as t h e power-cutoff c r i t e r i o n .
The burnout
For the
Under some conditions , tube-wall temperature o s c i l l a t i o n s of very
high amplitude were observed i n the low-flux b o i l e r . These f luc tua t ions
pers i s ted over long periods and had associated with them extreme var ia -
t i o n s i n both flow and pressure. A t the same time, the l i q u i d flow a t
the pump e x i t w a s steady t o within t h e precis ion of t h e electromagnetic
flowmeter. A t y p i c a l temperature t r a c e recorded during one such o s c i l -
l a t o r y s i t u a t i o n i s shown by the upper curve i n Fig. 2 .4 . The frequency
10
63
U N C L A S S I F I E D ORNL-LR-DWG. 59167
5 4
TIME-
Fig. 2.3. Sequence of Tube-Wall Temperatures D u r i n g Approach to Burnout.
4
.
11
1700
LL 0
I- U II: w a 1500 2 W I-
U N C L A S S I F I E D ORNL-LR-DWG. 60070
1800
l3Oo 0 I 20 40 60 80 100 120 140
TIME, sec
Fig. 2.4 . Tube-Wall Temperature Oscillations Observed in Low-Flux Boiler.
.
12
of the f luc tua t ion w a s -0.05 t o 0.10 cps with an amplitude of i l 5 O t o
2200°F.
conditions a t approximately the same i n l e t mass flow and e x i t qua l i t y
(6%). The frequency of the o s c i l l a t i o n w a s s i g n i f i c a n t l y higher (of
t he order of 0.35 cps) with amplitudes of only about +l°F.
served t h a t high i n l e t subcooling ( i n excess of 40°F) ex i s t ed f o r the
runs with high-amplitude wall-tempe ra ture o s c i l l a t i o n s . Further, s t ab le
operation w a s possible only when vapor bubbles were present i n t h e l i q u i d
a t t h e b o i l e r entrance.
In cont ras t , t h e lower curve t r a c e s a p a t t e r n seen under s tab le
It w a s ob-
Heat Transfer Results
Data from experiments with both b o i l e r s a r e given i n Tables 2.1 and
2.2. As indicated, t h e Series A data (Table 2.1) correspond t o t h e high-
f lux , small-diameter b o i l e r . While experiments with t h e low-flux, la rge-
diameter b o i l e r were completed p r i o r t o t h i s report ing per iod (experi-
ments were terminated by a physical burnout i n the pump c e l l ) , t he per-
t i n e n t r e s u l t s have been included for completeness; these have been desig-
nated Series B and a re given i n Table 2.2. t he small-diameter tube w a s e s s e n t i a l l y as i n the low-flux experiments;
however, t he decreased flow a rea i n the high-flux b o i l e r r e su l t ed i n
appreciably l a r g e r mass f luxes ( l b / h r . f t 2 ) .
r e l a t i o n was i n a l l cases based on the enthalpy change of t h e f l u i d across
the b o i l e r . Generally, t h i s value w a s cons is ten t t o within f20$ w i t h the
heat flux as ca lcu la ted from the temperature gradient i n the copper disks
o r the e l e c t r i c a l input t o t h e heaters (corrected by p reca l ib ra t ion for
t he ex terna l heat l o s s ) . E x i t q u a l i t i e s i n both b o i l e r s ranged between
10 and 9@.
The mass flow ( lb /hr ) )through
The heat f l u x used f o r cor-
The c r i t i c a l heat-f lux values obtained i n these experiments a re
shown i n Fig. 2 .5 .
minated i n "burnout"; while i n Ser ies B, runs 6.2-6.5 ( s t ab le w a l l tempera- t u r e ) and runs 5.1-5.3 and 8.1-8.3 ( o s c i l l a t i n g w a l l temperature) ended
with t h e reaching of t h e c r i t i c a l heat f l ux .
flow a r b i t r a r i l y se l ec t ed as an operat ional procedure w a s o f ten t o o la rge
I n Series A, runs 12-14, 16-22, 30-33, and 36-38 cu l -
The -16 reduction i n mass
.
t o sharply define t h e c r i t i c a l heat f l u x . For example, i n Ser ies A, runs
16-22, A21 showed s igns of i nc ip i en t burnout, while A22 had obviously
I
Tahle 2.1. Potassium Boiling: Ser ies A Results.*
I . .
io-= x Heat Balance Inlet Inlet Fluid Exit Fluid Exi t Boiler Heat Flux
A€' 'electric %hru copper ' %at R u n No. Mass Flow Temperature Temperature Qual i ty
Qfluid R,i - Tsat) (lb/hr * f t2 ) (OF) (OF) (vt $ v ~ r ) (Psi) (Btu/hr.ft2) Q,,,,
1 1.58 1422.0 1 3 9 . 8 10.9 2.7 60,200 0.96 0.72 7.5 2 1.49 1428.8 1389.7 17.9 3.4 94,000 0.97 0.79 11.0 3 1.21 1423.6 1388.0 21.4 2.8 92,300 0.98 0.80 11.1
4 1.51 1448.7 1402.1 21.1 4.4 112,000 1.18 0.90 19.0 5 1-33 1427.7 1 3 9 . 2 23.1 3.3 107,600 1.01 0.93 16.3
6 1.09 1425.3 1395.6 27.9 2.8 99,900 1.08 1.01 15.3
8 0.99 1439.4 1409.2 33.4 2.8 108,600 1.02 0.95 12.2
1.41 1443.9 1373.9 34.3 6.8 155,600 0.99 0'. 94 27.6 9 10 1.13 1438.5 1373 - 3 43.2 6.1 158,000 0.98 0.93 24.1 c3
7 1.23 1434.7 1404.6 27.1' 3.4 109,400 1.09 0.94 15 .1
P
11 0.99 1436.3 1374.7 48.4 5.3 157,200 0.99 0.98 20.1
l2 1.40 1508.2 1428.4 50.4 7.1 245,800 1.09 1.01 28.6 13 1.21 1494.6 1413.7 58.3 6.4 247,500 1.10 0.99 33.0 14 1.08 1410* 67.1 6.4 253,000 1.07 15 1.35 1385* 60.7 9.8 266,800 1.22
16 1.44 1469.7 1370 5 47.8 8.9 220,300 1.14 1.16 44.6 17 1.25 1463.0 1370.8 54.0 7.9 218,200 1.15 1.18 42.0 18 1.17 1458.2 1366.6 58.2 7.7 221, 100 1.14 1.16 40.1 19 1.04 1451.9 i366.6 64.8 6.9 =o,- 1.14 1.12 39-2 20 0.90 1453.2 1360.6 75.6 6.3 222,700 1.12 1.18 40.2
21 0.87 1362* 77.7 6.1 220,500*
22 0.77 1360* 86.2 5.7 220,500'
9 Table 2.1. Potassium Boiling: Series A Results (continued)
10'5 x Heat Balance Inlet Inlet Fluid Exit Fluid Exit Boiler Heat Flux
a/A 'electric 'thru copper msat Run No. Mass Flow Temperature Tempel'8tW Qual i ty
- Tsat) ( lb /hr - ft2) (OF) ( W (d % vapor) (Psi) (Btu/hr-f i2) % h i d
23 24 25 26 27
28 29 30 31 32
1.421 1.126 1.143 1.226 1.782
1.578 1.844 2.115 1.467 1.147
1424.4 1418.9 1448.6 1442.6 1461.1
1460.7 1466.4 1479.0 1472.1 1465.9
13%. 5 13%. 8 1403.0
1382.3
1386.5 1386.7 1397.3 1402.0 1402.6
1401.6
19.4 24.3 31.7 37.2 35.4
40.5 36.8 32.2 46.5 58.7
3.9 106,920 3.6 107,740 5.4 145,690 4.8 146,200
7.8 201,600 8.9 292,590 9.6 290,020 7.7 297,350 6.7 296,430
8.6 197, 220
1.l.20 1.028 1.080 1.086 1.117
1.0% 1.085 1.097 1.095 1 * 0953
0.8588 14.34 0.7853 12. so 0.979 19.57 1.0691 16.35 1.0638 39.68
1.0821 35.75 0.9834 40.72 1.0044 41.12 P
36.43 1.0201 1.0177 34.77
33 0.9367 1460.0* 1402. o* 73.4 6.5 304,960 1.0647
36 . 2.05 1481. g 1401. g 34.2 8.3 343,200 1.06 0.98 45.2 35 2.06 1447.4 1415.7 17.1 3.7 174,000 1.000 0.92
37 1.54 1463.7 1393 * 5 45.5 6.9 349,500 1.04 0.98 39.8 38 1.27 1450.6 1388.0 53.8 6.1 345,100 1.05 0.98 47.1
39 1.20 1462.4 1395.8 57.5 4.9. 347,100 1.05 0.98 55.5 40 2.08 1484.1 1405.3 34.2 8.7 346,900 . 1.06 0.99 43.4 4 1 1.81 1472.8 1394.4 39.9 7.7 357,000 1.03 0.95 41.1
%igh-flux boiler (0.325-in.-ID). *Est Fmated.
8
Table 2.2. Potassium Boiling: Series B Results.*
Heat Balance 10-5 x ‘electric %hru copper ms*t Inlet In le t Fluid E x i t Fluid E x i t Boi ler Heat Flux
Run No. Mass Flow Temperature Temperature Quality AP ‘/A ( m i d 8 (lb/hr.ft2) (OF) (OF) (wt vapor) (psi) (Btuhr-f t ‘fluid ‘fluid cTw, i - Tsat )
1335 *7 1328.4
* 5 -1, 0.28 1321.3 5 -2, 0.25 5.3 0.22
6.2 0.37 6.3 0-37 6.4 0.21 6.5 0.15
7.1 0.29 7-2 0.29 7.3 0.28
8.1, 0.23 8.2, 0.15 8.3 0.13
*
1401.5 1404.3 1378.1 1361.2
1373.0 1365.5 1395.1
1200.9 1215 .o 1207.0
1391.9 1396.0 1386.0
1404.7 1390.9 1354.6 1363.6
1348.7 1365 -3
1312.2
1369.5
1307.2
1300.9
42.4 45.1 50.0
32.5 34.3 60.8 81.3
48.9 49.5 51.2
33.0 50.9 61 .o
32,080 32,360 30,750
1 .9 31 890 1.22 2.0 32,970 1.18 1.8 33,060 1.18 1.6 32,370 1.21
2.2 37,630 1.18 2.5 37,630 1.19 2.3 37,970 1.18
21,900 21,000 21 $100
11.9 I-‘ VI 13.3
15.7 13.8
22.7 26.0 22.1
‘Low-flux bo i l e r (0.87-in.-ID). * Runs w i t h high-amplitude tube-wall temperature osc i l la t ions .
16
UNCLASSIFIED ORNL-LR-DWG. 74092 R !
4 x (05 2 4 6 8105
MASS FLOW, Ib/hr.ft2
Fig. 2.5. Critical Heat Flux with Boiling Potassium.
.
exceeded the l i m i t ; both poin ts a r e p l o t t e d i n Fig. 2.5. Comparison i s
made f o r t he da ta of each bo i l e r t o an ex t rapola t ion of t h e r e s u l t s of
Lowdermilk, Lanzo, and Siege15 f o r the forced-convection sa tu ra t ion b o i l -
ing of water,
as adjusted t o the spec i f i c geometries of t h e b o i l e r s . In Eq. 2.1,
(q/A)bo i s the c r i t i c a l heat f lux (Btu/hr . f t2) ; d, t he b o i l e r diameter
( f t ) ; L, the b o i l e r heated length (ft); and G, t he mass flux ( lbm/hr . f t2) .
For runs through A29 t h e length of t h e high f lux region (expressed i n
L/d) was 71; f o r runs A30 through 1141, t h e heated length w a s 45. The
t o t a l b o i l e r length w a s 1-35 L/d. The agreement between the potassium
experimental values and those predicted from water experience is seen t o
be qui te good. Since the thermal proper t ies of water and potassium a re
not too d iss imi la r , t h i s r e s u l t is not unexpected. Where s t ab le w a l l -
temperature f luc tua t ions of high amplitude exis ted, the c r i t i c a l heat
f l ux was reduced by about 4%; t h i s i s indicated by t h e dashed l i n e i n
Fig. 2.5.
with the values of 85 t o 95% observed when s t ab le operation ex i s t ed .
Data f o r heat f luxes below the c r i t i c a l are shown i n Fig. 2 .6 as a
function of a mean superheat based on a l i n e a r fluid-temperature va r i a -
t i ~ n . ~
sec t ion makes t h i s a reasonable assumption. For t h e high-flux bo i l e r ,
t he choice of a s t r a i g h t - l i n e fluid-temperature path is somewhat more
questionable. However, i n view of uncer ta in t ies i n the ca lcu la ted inside
w a l l temperature, no allowance w a s made f o r possible curvature; it w a s
estimated that t h i s could introduce an e r r o r of as much as 5% i n msat. The dispers ion i n HSat on any f l u x l eve l , which o r ig ina l ly appeared t o
r e l a t e systematical ly t o the e x i t qua l i ty , now seem t o be more random.
Some separat ion ex i s t ed i n the copper-s ta inless bond f o r runs following
A22; t h i s , cospled with the loss o f ' a s ign i f i can t number of thermocouples,
The e x i t qua l i t y with t h i s condi t ion ( -36) i s t o be contrasted
For the low-flux bo i l e r , t h e small pressure l o s s across the t e s t
~~ ~
5W. H. Lowdermilk, C . D. Lanzo, and B. L. Siegel, "Invest igat ion of Boiling Burnout and Flow S t a b i l i t y f o r Water Flowing i n Tubes, " NACA-TN- 4382, September 1958.
.
18
7 x
9 0
X 3 J LL
c U W I
2 x
UNCLASS I FlED ORNL-LR-DWG. 74091
1 o5 6 5 4
3
2 - 0.929-in. ID
I o5 8 6 5 4
3
1 o4 4 6 8 10 20 40 '6060100 200 400
TEMPERATURE DIFFERENCE, Tw,i-Tsat (OF)
Fig. 2.6. Boiling Potassium Heat Transfer.
0
19
may account f o r the lack of consistency. Results obtained by Brooks'
using a concentric double-tube b o i l e r w i t h heat t r a n s f e r from sodium t o
potassium have been included for comparison.
Pressure -Drop Data
The pressure l o s s across the b o i l e r i s shown i n Fig. 2.7 as the
var ia t ion of the pressure-loss r a t i o ( t o t a l two-phase t o equivalent
l iquid-only a t the same i n l e t mass flow) with t h e e x i t qua l i ty expressed
as t h e Lockhart-Martinelli7 parameter; a consis tent pa t te rn , with the
t o t a l two-phase pressure drop increasing as the e x i t qua l i ty increased,
w a s observed. In viewing t h i s r e s u l t , it should be kept i n mind t h a t the
pressure taps were located approximately 2 ft upstream of the b o i l e r en-
t rance and about 1 f t beyond the e x i t ; and thus, the data include t h e
(unknown) losses associated with flow through an i n l e t mixing chamber,
around exit-end thermowells, and through an expansion a t the e x i t .
obtained f o r the forced-convection boi l ing of water,. when presented i n
similar form but with t h e exclusion of losses incurred outside the con-
f i n e s of the boi le r , f a l l about a f a c t o r of th ree higher.
parison between potassium and water would be i n terms of the f r i c t i o n a l
port ion only of t h e pressure drop; however, information on the s l i p r a t i o
i n a mixed potassium liquid-vapor flow needed t o account f o r the accelera-
t i o n contr ibut ion t o t h e t o t a l momentum change i s not now avai lable .
Data
A b e t t e r com-
Future Experiments
A new b o i l e r , similar i n e s s e n t i a 1 , d e t a i l t o the high-flux sect ion
shown i n Fig. 2.2 i s being constructed.
Rh thermocouples f o r improved r e l i a b i l i t y and w i l l include three posi t ions
along the tube a t which w a l l temperatures w i l l be measured d i r e c t l y .
sure taps w i l l be located a t the i n l e t and e x i t and a t t h e midlength of
the b o i l e r .
This u n i t w i l l contain Pt-Pt 10%
Pres-
6R. D. Brooks, "Alkali Metals Boiling and Condensing Investigations, I'
Quar. Prog. Rep. Jan. 1 - Mar. 31, 1962, General Elec t r ic Co., F l igh t Propulsion Laboratory, Cincinnati , Ohio.
7R. W . Lockhart and R . C . Mart inel l i , "Proposed Correlation of Data f o r Isothermal Two-Phase, Two-Component Flow i n Pipes," Chem. Eng. Prog., 45: 39-48 (1949).
30
2 0 1/2
15
10
UNCLASSIFIED ORNL-LR-DWQ. 12 123
8 10 12 14 16 18 20 2 2 24 26 X10'2 0 2 4 6 8
2 0 9 P 0 5 P 0.1 LOCKHART-MARTINELL1 PARAMETER, X,, = (lY ) * (e) (e)
Fig. 2.7. Pressu re Drop with Flow of Boiling Potassium Through a 0.325-in.-diam Tube.
, I
21
Following completion of a s e r i e s of reca l ibra t ion experiments, the
current b o i l e r w i l l be removed; and the loop prepared f o r the i n s t a l l a -
t i o n of the new tes t sec t ion .
2.1.2. Superheat Phenomena
A. I. Krakoviak
' Analysis
A continued examination of the data obtained i n the forced-convection
loop as discussed above, along with t h a t reported on t h e performance of
a number of other n a t u r a l c i r c u l a t i o n and forced-convection systems used
i n mater ia ls compat ibi l i ty s tud ies a t ORNL, has l e d t o a de ta i led con-
s idera t ion of the r o l e of l i q u i d superheat i n the boi l ing process with
a l k a l i l i q u i d metals.' For example, i n one small natural-convection ap-
paratus with potassium, a boi l ing i n s t a b i l i t y of a cyc l ic nature w a s
noted.'
the b o i l e r increased from - 1 6 0 0 ~ ~ t o - 1 6 9 0 ~ ~ over a period of 40 t o 60 seconds; during t h i s time the vapor temperature decreased. The f l u i d then
erupted o r exploded, and both l i q u i d and vapor temperatures abrupt ly ap-
proached the value (1600 '~) corresponding t o the normal boi l ing tempera-
t u r e f o r the conditions of t h i s loop; t h i s process would then repeat i n
a cycl ic fashion. It would appear then t h a t s u b s t a n t i a l l i q u i d superheat
i s required t o i n i t i a t e bubble formation and t o s u s t a i n boi l ing .
can be demonstrated a n a l y t i c a l l y as follows :
Temperature records show t h a t the temperature of the l i q u i d i n
This
Consider first the condition f o r equilibrium of a spherical vapor
bubble i n a l i q u i d pool; t h i s can be expressed as,
P v - P Q = 2 0 r ,
'A. I. Krakoviak, "Notes on the Liquid Metal Boiling Process,"
'E. E . Hoffman ( i n discussion of paper by A. I. Krakoviak), "Pro-
USAEC Report ORNL-TM-618, Oak Ridge National Laboratory, 1963.
ceedings of 1963 High-Temperature Liquid-Metal Heat Transfer Technology Conference, 'I USAM: Report O R N L - ~ ~ O ~ , Oak Ridge National Laboratory, ( i n preparation) .
22
where Pv is sure i n the
the surface
the pressure of t he vapor within t h e bubble; Pa, t h e pres -
l i q u i d e x t e r n a l . t o the bubble; r, t h e bubble radius, and 0,
tension of t he l i q u i d a t the vapor-liquid in t e r f ace . For
bubble growth, Pv must be g rea t e r than Pa. bubble a l s o e x i s t s , t h e temperature of t he l i q u i d exceeds t h a t cor res -
ponding t o t h e sa tu ra t ion temperature of t h e l i q u i d a t pressure, Pv.
This excess temperature, over t h e sa tu ra t ion value a t Pa, i s termed the
l i q u i d superheat.
When evaporation i n t o the
If Eq. 2.2 is
by Ellion," there
combined with the Clausis -Clapyron r e l a t ion , as done
r e s u l t s t he expression:
*
wherein it has been assumed t h a t t he spec i f i c volume of t he l i q u i d can
be neglected i n comparison with t h a t of t h e vapor and t h a t t h e gas law
re l a t ion , W = RT, descr ibes t h e spec i f i c volume of t h e vapor. I n Eq.
2.3, Tw i s the temperature of t h e surface on which t h e bubble forms; , t he absolute temperature of t he sa tu ra t ed l i q u i d a t P R, t h e gas Tsat J a ;
constant ; and h t h e l a t e n t heat of vaporizat ion of t h e l i q u i d . This fg'
equation then p red ic t s t he approximate l i q u i d superheat required f o r
equilibrium f o r a bubble of radius, r .
Equation 2.3 has been t e s t e d and found adequate by a number of ex-
perimenters. G r i f f i t h and Wallis" s tudied nucleat ion of water, ethanol,
and methanol from a copper surface containing conica l c a v i t i e s of 0.0018- i n . diameter a t t h e base. With t h i s value f o r r, the ca l cu la t ed l i q u i d
superheat agreed wel l with t h e experimental r e s u l t s f o r t h e superheat r e -
quired t o maintain e b u l l i t i o n . Berenson12 obtained s i m i l a r r e s u l t s i n
. .
'%I. Ell ion, "Study of Mechanism of Boi l ing Heat Transfer ," Jet
"P. G r i f f i t h and J. D. Wallis, "The Role of Surface Conditions i n
12P. Berenson, "Transit ion Boiling Heat Transfer from a Horizontal
PropulsioE Laboratory Memorandum 20-88, March 1954.
Nucleate Boiling, Chemical Engineering Symposium Ser ies , 56: 49-63 (1960).
Surface," MIT Technical Report N o . 17, Heat Transfer Laboratory, Massachusetts I n s t i t u t e of Technology, March 1960.
'
c 23
.
h i s inves t iga t ion of t h e e f f e c t s of surface roughness on the superheat
involved i n the bo i l ing of l i q u i d pentane. . .
Since Eq. 2.3 appears t o represent t h e r e s u l t s fo r water and severa l
organic f l u i d s , it is presumed t h a t t h e equation may a l s o be appl icable
with l i q u i d metals. For ease i n ca lcu la t ion , t h e evaluat ion of t h e l i q u i d
superheats required for boi l ing with the severa l l i q u i d metals has been
made i n comparison with water. Thus, using Eq. 2.3,
1 - ( T w . i Tsat w
1 - ( T w , i 'sat x
where the subscr ip ts , w and x, denote respec t ive ly water and l i q u i d metal,
and V
the vapor volume).
bubble formation from a p a r t i c u l a r surface with c a v i t i e s of radius, r,
immersed i n water at atmospheric pressure, the superheats with t h e l i p i d
metals were determined t o have the values given i n Table 2.3.
is t h e spec i f i c volume change between l i q u i d and vapor (E V fg V'
Taking 30°F as the superheat required t o sus t a in
Table 2.3. Estimated Liquid Superheats Required t o Sustain Boiling with
the Alka l i Liquid Metals
F lu id Superheat, "F
N a
K
R b
cs
One fu r the r f a c t o r may be considered i n r e l a t i o n t o the superheat
problem. The bo i l ing s i t e s ( c a v i t i e s ) ava i lab le on a na tura l surface
(such as a pipe w a l l ) w i l l show a normal d i s t r i b u t i o n of s i z e s . From Eq. 2.3, it i s seen t h a t as the temperature of t he surface r i s e s , bubble
formation w i l l begin f i rs t i n the c a v i t i e s of l a r g e s t radius .
s i t e s are used up, t he w a l l temperature w i l l continue t o r i s e and smaller
and smaller c a v i t i e s w i l l be brought i n t o ac t ion . Presumably these
As these
24
.
c a v i t i e s are not wet by the f l u i d ; i . e . , t he f l u i d does not displace
trapped or adsorbed gases. On the o ther hand, l i q u i d metals wet most
metals e a s i l y . It i s t o be expected then t h a t the bo i l ing s i t e d i s -
t r i bu t ions w i l l be d r a s t i c a l l y skewed, predominating i n very small cavi-
t i e s . The superheats required t o i n i t i a t e bubble formation with the
l i q u i d metal w i l l thus s i g n i f i c a n t l y exceed t h a t required with water.
In addi t ion, s ince the l i q u i d metals used a re general ly of very high
p u r i t y and a re i n contact with clean surfaces i n the absence of i n e r t
gases, the comparison of Eq. 2.4 should be made with water under similar
condi t ions. It has been observed t h a t f o r d i s t i l l e d , deaerated water
i n contact with a very clean surface the l i q u i d superheat may be as high
as 90°F. Thus, t he values given i n Table 2 .3 may be as much as a f a c t o r
of th ree g rea t e r .
The thermal d i f f i s i v i t i e s and t h e thermal conduct iv i t ies of t h e
l i q u i d a l k a l i metals grossly exceed those f o r water; e .g . , t he thermal
conductivity of potassium i s near ly 50 times t h a t f o r water, and t h e
thermal d i f f u s i v i t y r a t i o , K t o Ha0 is of t he order of 335. Thus con-
s ider ing conduction only, the r a d i a l temperature d i s t r ibu t ions i n a
s t a t i c column of water and potassium w i l l be s i g n i f i c a n t l y d i f f e r e n t ;
a comparison i s made i n Fig. 2.8 f o r the condi t ion of a uniform hea t
input of 30,000 Btu/hr . f t2 through t h e w a l l of a c i r c u l a r tube of l - i n .
diameter. With water, a bubble nucleat ing a t the w a l l grows rap id ly
i n the t h i n f i l m of superheated l i q u i d near t he surface (-0.005 i n . as
indicated i n Fig. 2.8b) and e i t h e r col lapses a f t e r leaving the w a l l ,
i f t h e bulk f l u i d i s subcooledj or remains as a s t ab le o r a slowly grow-
ing e n t i t y , i f t h e bulk f l u i d is sa tura ted . I n con t r a s t , with potassium,
the bubble (once formed) remains i n a highly superkeated environment even
at a dis tance f a r from t he w a l l and can be expected t o grow rap id ly
(even explosively) u n t i l t he t o t a l energy ava i lab le from the superheat
i s consumed. The energy ava i lab le p e r u n i t volume i n such a s i t u a t i o n
(assuming again a tube of l - i n . diameter f o r which the cen te r l ine tem-
perature i s -340'F above sa tu ra t ion ) i s s u f f i c i e n t t o vaporize the
potassium t o a qua l i t y of about 7.5%. void f r ac t ion of 0.99 (assuming zero s l i p ) .
This qua l i t y corresponds t o a
The mechanism pic tured
L . .
UNCLASSIFIED ORNL- DWG 63-802
2000
G+A-
1500
!i?
a
a
- W
3 t 1000
5 W a
c
500
G+A-
0
\ b I
-0.5 0 0.5 RADIUS (in.)
( 0 )
40
30
A
k
a g 20
Y
l-
w
W a 3 m
3
-I
P 0
f0
0
I WATER TEMPERATURE PROFILE NEAR WALL
400
300
h
5 a Y
I-
w I E 200 a
0 0
3 v)
3
J
1 00
0
POTASSIUM w TEMPERATURE PRI 'ILE TO CENTER
OF TUBE
I 0 0.005 0.010 0 0.1 0.2 0.3 0.4 0.5
DISTANCE FROM HEATED SURFACE (in.) DISTANCE FROM HEATED SURFACE (in.)
(6) (C)
Fig. 2 . 8 . Comparison of Radial Temperature Distributions in Water and Fotassium Due to Conduction. q / A = 30,000 Btu/hr.ft2; D = 1 in.
26
i s cons is ten t with the pa t t e rn of temperature o s c i l l a t i o n s observed i n
the low-flux b o i l e r during the previously described periods of high-
amplitude tube - w a l l temperature f luc tua t ions . 4, ' I
As noted e a r l i e r , s t ab le performance ( i . e . , e s s e n t i a l l y s teady tube-
w a l l temperatures) has been observed only when vapor ex i s t ed a t the
b o i l e r i n l e t . This vapor w a s generated e i t h e r by f lash ing across the
i n l e t flow r e s t r i c t i o n o r by allowing some bo i l ing i n the i n l e t mixing
chamber; the i n l e t qua l i t y has been estimated a t -0.5 w t %. s i b l e then t h a t s t a b i l i t y resu l ted from the presence of the bubbles i n
the l i qu id , thus circumventing the high superheat necessary f o r bubble
nucleation. A t the same time, t he da ta of Radovcich and Moissis14 f o r
the flow of air-water mixtures i n a v e r t i c a l tube suggest t h a t an annular
flow w i l l e x i s t f o r t h e conditions of t h i s experiment (potassium flowing
a t G = 0.6 x lo5 l b / h r . f t 2 i n 0.325-in. diameter tube with 0.5% i n l e t
q u a l i t y ) .
l a t e d t o be *0.040-in., assuming zero s l i p .
conservative superheat value given i n Table 2 .3 f o r potassium, conduction
alone w i l l s u s t a in a thermal f lux through a potassium f i l m of 0.040-in.
thickness of near ly 0.67 X lo6 Btu/hr.ft" without nucleat ion.
a l l experiments t o date , the heat t r ans fe r r ed during s t ab le operation
can be accounted f o r by conduction through a t h i n l i q u i d f i lm on the tube w a l l followed by evaporation a t the in te r face between t h e l i q u i d
f i l m and the c e n t r a l gaseous core.
It is pos-
The thickness of t he l i q u i d f i lm on the w a l l has been calcu-
Thus, allowing even the
Hence, i n
Ekpe rime n t
Experiments t o quant i ta t ive ly define the above phenomena and hope-
f u l l y t o ve r i fy t h e predicted superheats have been i n i t i a t e d .
na tura l c i r cu la t ion loop w i l l be used.
t u s i s s t i l l i n a preliminary s tage, f u r t h e r descr ip t ion w i l l be post-
poned t o a l a t e r r epor t .
A small
Since t h e design of t h i s appara-
'9 rev ious Section 2.1.1, Operating Charac t e r i s t i c s .
I%. A . Radovcich and R . Moissis, "The Trans i t ion from Two Phase Bubble Flow t o Snug Flow," MIT Report No. 7-7673-22, Department of Mechanical Engineering, Massachusetts I n s t i t u t e of Technology, June 1962.
27
2.2. Water Boiling i n a Multirod Geometry
J. L. Wantland
The design optimization of a multirod f u e l element cooled by a
bo i l ing l i q u i d flowing a x i a l l y along t h e rod c l u s t e r requires a knowl-
edge of t h e maximum allowable (burnout) heat f l ux and of the two-phase
pressure drop as these f ac to r s a re influenced by the element geometry.
The problem of burnout, i n p a r t i c u l a r , for such complex geometries has
received considerable a t t e n t i o n during the pas t few years .
general ly c o n ~ l u d e d ~ ~ ' ~ ~ from s tudies with i n t e r n a l l y heated annul i
and with 3-, 7-, and 19-rod c l u s t e r s t h a t the burnout heat f l ux is
lower than f o r s ing le round ducts a t the same conditions of qua l i ty ,
mass veloci ty , and pressure. a t t he same heat f l ux and s t a t i c pressure, t h e burnout steam q u a l i t y
w a s of t he order of 75% i n a round tube of 10 rnm diameter and only
20% i n a 3-rod c l u s t e r a t 6 rnm separat ion.
difference have not been completely es tab l i shed , data on the f i l m t h i ck -
ness with the flow of a i r -water mixtures i n an annulus2' indicate a
th inner f i lm on the c e n t r a l rod than on t h e tube wal l . Becker concludes
It has been
Becker," f o r example , ind ica tes that
While t h e causes of t h i s
" S . Levy, E. E . Polomik, C . L. Swan, and A. W . McKinney, "Eccentric Rod Burnout a t 1000 lb f / in .2 with Net Steam Generation, It In t . J. Heat Mass Transfer, 5 : 595-614 (Ju ly 1962).
"K. M. Becker, "Burnout Conditions f o r Flow of Boiling Water i n Ver t ica l Rod Clus te rs , " A.1.Ch.E. Journal, 9: 216-222 (March 1963).
17f'Basic Experimental Studies on Boiling F lu id Flow and Heat Trans- f e r at Elevat'ed Pressures , '' Task X I 1 1 Progress Reports MRP-x-3-61 through M P R - x - ~ - ~ ~ , Columbia University (1961-1962).
18G. M. Hesson e t a l . , "Preliminary Boiling Burnout Experiments with a 19-Rod Bundle Geometry i n Axial Flow," USAEC Report HW-73395, Hanford Laboratories, 1962.
19.J. G. Co l l i e r , "Heat Transfer and Flu id Dynamic Research as Ap- p l i e s t c Fog Cooled Power Reactors , ' I Atomic Energy of Canada Limited Report CRARE-1108 , June 1962.
*'G. F. Hewitt and A. T . G . Stevens, "Two-Phase Annular Flow of Air/Water Mixtures i n , an Annulus. Drop Measurements over a Range of Phase Flow Rates," B r i t i s h Atomic Energy Authority Report AERE-R-3957.
Par t I: Film Thickness and Pressure
from h i s study t h a t t h e r a t i o (7) of the heated perimeter t o the t o t a l
wetted perimeter is a s ign i f i can t var iab le and t h a t as the number of
heated elements i n the c l u s t e r increases ( i i e . , as 7 + 1.0) the burnout
qua l i t y should approach t h a t f o r flow ins ide tubes.
There e x i s t s , however, s ign i f i can t disagreement between var ious in-
ves t iga tors as t o the physical loca t ion of t h e burnout region. Thus,
Levy e t a1.,15 studying an eccent r ic annulus, repor t burnogt on t h e heated
c e n t r a l rod i n t h e region of c l o s e s t approach t o t h e outer w a l l of t h e
annulus. 1-b was a l s o observed t h a t , a t t h e smallest gap, t h e burnout
heat f l ux w a s redcced 15' t o 3@ below the concentric annulus value.
cont ras t , Becker" w i t h a 3-rod c l u s t e r fourid t h a t burnout conditions ex-
i s ted on the por t ion of t h e rods facing t h e l e a s t r e s t r i c t e d flow areas
and t h a t t he re w a s no s ign i f i can t e f f e c t of rod spacing.
show t h e absolute difference i n qua i i t y a t burnout between a 6 m and a
2 rn spacing t o be 7% (the percentage change is about 25%). It is not
a t a l l c l e a r t h a t t h i s should be considered in s ign i f i can t . 1 The inves t i -
gat ions of Becker and Levy d5f fer pr imar i ly i n (1) the pressure l e v e l and
(2) t he shape, area, ar,d boundary conditions f o r t he flow channel sur-
rounding a u n i t rod. Both of these f ac to r s could be of importance i n
explaining t h e discrepancies between the two experiments.
In
[Becker's da ta
,
With the goal cf resolving some of these uncer ta in t ies , an experi-
mental Invest igat ion of t he phenomena associated w i t h b o i l i z g i n multirod
geometries has been i n i t i a t s d . The e f f e c t s of rod spacing, i n l e t sub-
c o d i n g , and o u t l e t quality on the burnout heat f l ux w i l l be considered
over a range of near atmospheric pressure using deaerated, demineralized
water as the t e s t f l u i d . In a l a t e r aspect of t h i s study, an apparatus
w i l l be c o n s t n c t e d t o examine bo i l ing w i t h potassium i n a 7-rod geometry.
A photograph of t he experimental system assembled f o r these s tud ies
i s given i n Fig. 2.9. The outer s h e l l s of a l l major loop components have
been fabr ica ted from flanged s e c t i o m of g l a s s piping t o allow v i sua l ob-
se rva t ion of phenomena occurring during bo i l ing runs; a t t h e same time,
easy removal and exchange of individual u n i t s w i l l be poss ib le .
b o i l e r w i l l contain a c l u s t e r of seven 1/2-in. diameter Firerods of 10-in.
heated length arranged i n a t r i a n g u l a r a r r a y with six rods surrounding
The
-
.
. I
~* 29
.
UNCLASSIFIED
.
Fig. 2.9. View of Apparatus fo r Study of C r i t i c a l Heat Flux with Water Boiling i n a Multirod Channel.
the c e n t r a l seventh rod.
w i l l be possible . Thermocouples embedded i n the rods near the surface
w i l l provide both information on surface temperatures and the input s i g -
Surface heat f luxes t o 1.6 X lo5 Btu/hr.ft2
na l t o a device f o r protect ion against a c t u a l physical burnout.
c l u s t e r s with rods a t spacings of 1/16, 1/8, and 3/16 i n . i n channels of
1 3/4-, 2-, and 2 1/4-in. diameter, respectively, a re being fabr ica ted
f o r the i n i t i a l s tud ies ; the corresponding r a t i o s of heated-to- total pe-
rimeter a re 0.67, 0.64, and 0.61. Dummy rod sect ions may be inser ted i n
the channel t o simulate hydrodynamic conditions of an i n f i n i t e a r ray of
rods. The e f f e c t s of roughness elements attached t o the channel w a l l on
the burnout f lux w i l l a l s o be considered. Operation under both natural-
and forced-circulat ion conditions w i l l be possible .
Boiler
2.3. General Boiling Studies
W . R . Gambill
2.3.1. Predict ion of C r i t i c a l Heat Fluxes
The f a c t t h a t a boi l ing system i s characterized by a peak of c r i t i -
c a l heat f lux f o r which t r a n s i t i o n occurs from e f f i c i e n t nucleate boi l ing
t o high-thermal-resistance film boi l ing has prompted the proposal of many
equations f o r predict ing the peak f l u x . For a constant heat-input device
such as a nuclear reactor , such a predict ion is of prime importance; since
the rapid r i s e of surface temperature which follows attainment of t h e c r i t -
i c a l heat f lux usual ly r e s u l t s i n melting o r burnout of the heater sur- face; the term "burnout heat f lux" i s of ten used, as here, t o denote the
heat f lux a t the c r i t i c a l bo i l ing t r a n s i t i o n .
Although previous cor re la t ions have been useful f o r spec i f ic l imi ted
conditions, they are general ly characterized by one or more shortcomings
which considerably r e s t r i c t t h e i r general i ty :
1. The form of t h e equation may be such as t o pred ic t a c r i t i c a l .
f lux of zero when the subcooling i s zero, which i s contrary t o experience.
2 . The equation may be applicable t o only a s ingle f l u i d , usually
water.
.
..
.
3. The equation may have been s t a t i s t i c a l l y derived i n a fashion
far removed from physical considerations, an a r b i t r a r y attempt being made
t o r e l a t e t h e c r i t i c a l f lux t o the numerous experimentally cont ro l led
var iab les .
4. A given co r re l a t ion , when compared with data other than those
from which it w a s derived, i s of ten found simply t o give poor agreement with the o ther da ta .
A new addi t ive co r re l a t ion , general ly appl icable t o flowing, sub-
cooled, wetting l i qu ids , has been developed21 i n which a term represent-
ing the boi l ing contr ibut ion t o t h e c r i t i c a l heat flux i n the absence of
forced convection is added t o a term represent ing the equivalent forced-
convection cont r ibu t ion i n t h e absence of bo i l i ng . The predict ions based
on t h i s superposit ion method have been compared with a l l avai lable p e r t i -
nent experimental da ta (1389 t e s t s ) . These da ta a re f o r seven f l u i d s i n
tubular , annular, rectangular , and rod geometries over very broad ranges
of flow conditions:
p s i a ; subcooling, atsub, 0-506 '~ ; accelerat ion, a, 1 - 5 7 , O O O gee; and burn-
out heat f lux , 'bo, 0.01 X lo6 - 37.4 x lo6 Btu/hr . f t2 .
a re excluded which can be shown t o be unrepresentative of burnout under
the experimental condi t ions, a se lec ted data f i e l d of 941 t e s t s remains,
for which 9676 of the predict ions agree with the experimental values within
a maximum deviat ion of 46. are compared f o r a l l ava i lab le da ta f o r burnout with water coolant only
(943 t e s t s ) , it is found t h a t t he addi t ive co r re l a t ion gives a deviat ion -37% l e s s than Bernath's extended co r re l a t ion .
ve loc i ty , V, 0.005-174 f t / s e c ; pressure, P, 4.2-3000
When t'hose data
I f deviat ions which include 9 6 of the data
22
The co r re l a t ion may be summarized as follows:
21W. R . Gmbi l l , "Generalized Predict ion of Burnout Heat Flux f o r Flowing, Subcooled, Wetting Liquids, " Chemical Engineering Progress Symposium Series , 59(41): 71-87 (March 1963).
t o Exis t ing Data, I' Chemical Engineering Progress Symposium Ser ies , 22L. Bernath, "A Theory of Local Boiling Burnout and Its Application
56( 30) : 95-116 (1960).
32
where
. and
0.923
Fsub = 1 +
I n addi t ion:
I n these equations, $bo is the burnout heat f lux ; K, a constant (0.12
t o 0.17); k, t he l a t e n t heat of vaporizat ion of t he coolant ; p and a t he l i q u i d and vapor dens i t ies , respect ively; 0, t h e surface t en -
s ion; g , the force-to-mass conversion constant , - a, t h e t o t a l acce l -
e r a t ion ac t ing on t h e cooled surface; subscr ip ts ' b o i l ' and 'nb' denote
boi l ing and nonboiling, respect ively; c t h e constant-pressure spec i f i c
sub' heat ; A t
and tw and t
Equation 2.7 i s Kutateladze s empir ical dimensionless subcooling f a c t o r
f o r pool-boiling burnout as described by B ~ n i l l a . ~ ~ The w a l l tempera-
t u r e a t burnout, (tw)bo , is evaluated with Bernath's general ized plot22 of the l iqu id- f i lm superheat a t burnout as a funct ion of the reduced
sa tu ra t ion temperature.
%' C
P' the subcooling; h, t h e surface hea t - t ransfer coe f f i c i en t ,
t he w a l l and bulk coolant temperatures, respec t ive ly . b'
2.3.2. Randomness i n the C r i t i c a l Heat Flux24
I n an e f f o r t t o a sce r t a in experimentally the inherent uncer ta in ty
of the c r i t i c a l heat flux i n sa tura ted pool bo i l i ng (a matter which has
'3. F. Bonilla, pp. 399-431 of Chapter 9 ("Heat Removal") i n N-dclear Engineering, McGraw-Hill Co. , Inc . , New York, 1957.
2%. R. Gambill, "An Experimental Inves t iga t ion of t h e Inherent Uncertainty i n Pool-Boiling C r i t i c a l Heat Fluxes t o Saturated Water, t o be presented a t A.1.Ch.E. National Meeting, San Juan, Puerto Rico, September 1963.
33
c
.
been inconclusively debated f o r some t ime) , a s e r i e s of 234 t e s t s w a s
conducted w i t h d i s t i l l e d , deionized water outside horizontal , a-c heated,
A-nickel tubes i n a pool 6 x 6 x 9-in. deep. The maximum r e l a t i v e uncer-
t a i n t y i n t h e derived c r i t i c a l heat f lux values w a s +3$.
Six t e s t s were conducted with tubes which were taken t o physical
burnout, and -50 addi t ional t e s t s were made with each of four tubes pro-
t e c t e d by a burnout detector c i r c u i t which terminated the applied current
before the w a l l temperature exceeded ~450'F. r ized as follows ( for sa tura ted water a t an average pressure of 14.4 p s i a ) :
The r e s u l t s may be summa-
Table 2.4. Summary of Results i n Study of Randomness i n C r i t i c a l Heat Flux
$crit, Btu/hr. f t 2
Range Average Tube No. No. Tests
1-6 6 0.368-0.546 0.464 7 49 0.334-0.548 0.472 8 50 0.483-0.596 0.537 9 . 52 0.413-0.498 0.457 10 54 0.299-0.450 0.353 10a 23 0.201-0.369 0.280
The range of heat f luxes given i n Table 2.4 correspond t o values of
' K ' i n t h e Kutateladze-Zuber burnout equation (Eq. 2 .6 when FSub = 1)
of 0.076 t o 0.224. The average temperature of the cooled surface a t the
c r i t i c a l f lux f o r a l l the t e s t s was 252'F, which corresponds t o an aver-
age c r i t i c a l superheat of 41°F.
Large c r i t i c a l f luxes were observed when a very t h i n dark deposit
o r coating became v i s i b l e on the tube surface, and i t s absence generally
corresponded t o smaller c r i t i c a l f luxes . The coating, when it formed,
d id not, a f f e c t the measured surface roughnesses, which were -14 pin. rms
(+3 maximum).
Tests made w i t h tube No.. 9. were conducted under conditions of minimum
surface v a r i a b i l i t y , and f o r t h i s t e s t sect ion, the m a x i m u m s c a t t e r i n
\
t h e c r i t i c a l f luxes (k9.346) w a s approximately t r i p l e t he maximum r e l a t i v e
experimental uncertainty. If t h e maximum r e l a t i v e experimental uncer-
t a i n t y i s taken-as equivalent t o +3 standard deviat ions when considering
tube N o . 9 only, it may be shown t h a t t h e p robab i l i t y of t h e exis tence
of an inherent uncer ta in ty i n the c r i t i c a l - f l u x phenomenon i s >99.5$, when t h e evaluation is made a t t h e 99.9546 confidence l e v e l .
It is concluded that (a) the re i s an inherent uncer ta in ty o r i n t r i n -
s i c randomness i n t h e c r i t i c a l f l u x determined under sa tura ted pool bo i l -
ing conditions which approximates t h a t predicted by Zuber's hydrodynamic
i n s t a b i l i t y theory,2s and (b) t h a t t h e nature and condi t ion of t he cooled
surface can cons t i t u t e an inrportant influence on the c r i t i c a l heat f l ux .
2.3.3. In t e rna l Heat Transfer i n a Horizontal
Tube Immersed i n a Water Pool
Pool heat - t ransfer s tud ies have previously been conducted with ex ter -
n a l l y and i n t e r n a l l y cooled v e r t i c a l tubes and w i t h ex t e rna l ly cooled
hor izonta l tubes, but no study of i n t e r n a l l y cooled hor izonta l tubes has
appeared.
ducted using ac -heated nickel tubes which were ex te rna l ly vacuum jacketed.
Since t h e ends of t h e tubes were open and most of t h e heat w a s d i rec ted
r a d i a l l y inward, a pulsa t ing flow condi t ion w a s obtained i n a majori ty of
t h e t e s t s , during which l i q u i d flow inward toward t h e tube cen te r a l t e r - . nated w i t h two-phase pulses of water and steam toward t h e ends. This
pu lsa t ing flow w a s c h a r a c t e r i s t i c of tubes of l a rge L/D r a t io ; and with
such tubes, t he w a l l temperatures increased gradually without sudden ex-
cursions as t h e heat f l ux w a s increased. With tubes of small L/D r a t i o ,
howeve r , behavior c h a r a c t e r i s t i c of c rit i c a l - f lwc attainment w a s observed,
a t which f lux the w a l l temperature rose rap id ly .
An exploratory inves t iga t ion of t h i s case w a s accordingly con-
A t o t a l of 1.71 t e s t s was conducted with water at atmospheric pressurg
using tube of 1/8-, 1/4-, and 3/8-in. outs ide diameter, nominal heated
lengths of 4, 8, and 12 i n . , and th ree degrees of pool subcooling. An
.
=N. Zuber and M. Tribus, "Further Remarks on the S t a b i l i t y of Boil- ing Heat Transfer, ' ' University of Ca l i fo rn ia Engineering Department Report 58-5, January 1958.
35
c
.
addi t iona l s e r i e s of 3 l t e s t s w a s conducted i n an air environment.
data are being reduced t o show the dependence of average and maximum w a l l
temperature and two-phase pulse r a t e on average heat f l u fo r various
diameters, lengths, and subcoolings.
The
2.4. Swirl-Flow Heat Transfer
W . R . Gambill
2.4.1. Heat-Transfer Charac te r i s t ics of Ethylene
In order t o gain a more complete understanding of the influence of
physical p roper t ies on swirl-flow heat t r a n s f e r , a swirl-flow study of
ethylene glycol was i n i t i a t e d fo r comparison with t h e r e s u l t s obtained
previously using water.
the pool boi l ing, forced-convection nonboiling and l o c a l boi l ing, and
burnout c h a r a c t e r i s t i c s of ethylene glycol i n both a x i a l and s w i r l flow.
A few t e s t s were made under cooling condi t ions.
The program w a s broadened t o f i n a l l y encompass.
Apparatus
The system used w a s bas i ca l ly the same as t h a t used f o r t he water
studies27 (Fig. 2.10).
together t o pump t h e glycol through horizontal copper o r 347 s t a i n l e s s
s t e e l t e s t sect ions, res i s tance heated by 60 cycles/sec ac cur ren t .
The glycol w a s contained i n a mixed overhead holdup drum which could be
steam heated o r w a t e r cooled'with i n t e r n a l c o i l s . A cooler tube, which
followed t h e hea ter tube i n some t e s t s , w a s ex te rna l ly jacketed t o form
an annular flow passage for longi tudina l flow of cooling water; 40-gauge
(3-mil-dim) chromel/alumel thermocouples were embedded i n the cooler
tube wal l .
used i n a l l axial-flow t e s t s .
Moyno and turb ine pumps were used s ingly or
An upstream calming sec t ion longer than 40 diameters w a s
Swirl flow w a s generated with t h i n
.
26W. R . Gambill and R . D. Bundy, "High-Flux Heat-Transfer Charac- t e r i s t i c s of Pure Ethylene Glycol i n Axial and S w i r l Flow," A.1.Ch.E. Journal, 9(1) : 55-59 (1-963).
and Pressure Drop f o r Water i n Swirl Flow Through Tubes with In t e rna l Twisted Tapes, I' U W C Report ORNL-29ll, Oak Ridge National Laboratory, 1960; see a l so , Chem. Eng. Prog. Symposium Series , 57(32): 127-137,
27W. R . Gambill, R . D. Bundy, and R . W. Wansbrough, "Heat, Burnout,
(1961 1.
36
UNCLASSIFIED ORNL-LR-DWG 29245A
TURBINE MOY NO PUMP PUMP UNTREATED
TAP WATER
I I I
MULTI-POINT 1 RECORDER L
r- I
-1
- :::.-x> <:KQi-;:>>:: TEST SECTION TO ATMOSPHERE
MIXING CHAMBER y
HIGH PRESSURE
ROTA M E TE R S
TO WEIGH TANK
375 kva 0.5 TO 2i.O v
I
DISTRIBUTION TRANSFORMER
~ ( O V - 6 6 0 V-
440 v AC INPUT
Fig. 2.10. Experimental Apparatus for' Vortex-Flow Heat-Transfer Studies.
37
(15-mil-thick) twisted tapes , onto which the tubes were drawn; a des-
c r i p t i o n of these tubes i s t o be found i n Ref. 27.
Burnout t e s t s were concluded by physical destruct ion of the t e s t
Burnout w a s a t t a i n e d by slowly sect ion (usually near t h e flow e x i t ) .
increasing t h e heat flux at e s s e n t i a l l y constant veloci ty , pressure,
and i n l e t glycol temperature.
The pcol b o i l e r measured nominally 6 x 6 x 9 i n . deep. The hori- zontal heater tube w a s s i l v e r soldered across two insulated copper end
p l a t e s which served as electrodes; the bottom and s ides were made of
welded s t a i n l e s s s t e e l p l a t e .
t e s t s ) were placed inside the heater tube and longi tudinal t raverses
made under heating conditions t o determine the locat ion of the isothermal
zone. External s i l v e r voltage leads and the four. i n t e r n a l thermocouples
were then located within t h i s zone.
Thermocouples (chromel/alumel i n a l l
Three analyses of the glycol i n the loop were made during the t e s t s ;
the composition ranged from 99.6 t o 99.8 w t % glycol (balance water) .
No glycol decomposition o r w a l l deposit ion w a s noted.
Nonboiling Heat Transfer - Axial Flow
The axial-flow data were cor re la ted t o within average and m a x i m u m
deviations of 7.3% and 17.8% respectively, by:
Npr represent the Nusselt, Reynolds, and Prandtl moduli, NNu’ NRe’ vhe re ,
respect ively.
i n evaluating the modulus; and t h e subscr ipt E, that t h i s is . a mean value.
The subscr ipt b - ind ica tes t h a t bulk propert ies a re used
Note t h a t a t the high Reynolds moduli and heat f luxes of t h i s experi-
ment, the Nusselt modulus dependence on the Reynolds modulus i s very close
t o the f i rs t power ra ther than the more commonly observed 0.8 power.
Nonboiling Heat T r a m f e r - Swirl Flow
Both the glycol and the e a r l i e r water27 swirl-flow heating data have
been cor re la ted t o within an average deviation of +12.0$ by the r e l a t i o n :
38
0.0344
(2.10) 1 = 0.00675 3 (NNu)sm
[ < N R e ) ~ ' " s
i n which the dimensional group' on the r i g h t s ide of t he equation(propor-
t i o n a l t o t h e buoyant force per u n i t f l u i d volume i n t h e boundary l a y e r )
should be evaluated i n u n i t s of lb/ft3, f%/sec, ft, and OF.
p is the f l u i d densi ty; B, t he volumetric coe f f i c i en t of thermal expansion
of t h e coolant; A t , the f i l m temperature difference; ua, the a x i a l coolant
ve loc i ty ; y, t he tape- twist r a t i o expressed as ins ide tube diameters per
180-deg t w i s t ; and Di, t h e i n t e r n a l tube diameter.
ca t e s s w i r l flow; g, t he mean value; and - a, t h a t t he Reynolds modulus
is based on the a x i a l flow ve loc i ty and t h e tube ins ide diameter.
In Eq. 2.10,
The subscr ip t ind i -
Burnout Heat Flux
The atmospheric pressure, sa tura ted , pool-boi l ing c r i t i c a l f l u x f o r
ethylene glycol w a s found t o be u168,000 Btu'/hr-ft2.
The forced-convection a x i a l - and swirl-flow burnout da ta obtained
are given i n Table 2.5. The predic t ions of t h e addi t ive burnout cor re la -
t ion21 are i n good general agreement with these da t a .
Cone l u s ions
It w a s concluded th6t:
1. The f r i c t i o n - f a c t o r equation as given i n Ref. 27 i s adequate
f o r pred ic t ing isothermal f r i c t i o n f ac to r s fo r a v a r i e t y of fluids i n
s w i r l flow.
2. Organic f l u i d s , and perhaps water, give l a r g e r nonboiling
ax-ial-flow heat t r a n s f e r coe f f i c i en t s at high v e l o c i t i e s and hea t f luxes
than predic ted by s tandard co r re l a t ions .
number exponent appears t o be nearer 1 .0 than 0.8.
For such condi t ions t h e Reynolds
3. Use of t he buoyant force pe r u n i t coolant volume gives a b e t t e r
ove ra l l co r re l a t ion of heating and cooling swirl-flow heat t r a n s f e r coef - f i c i e n t data than does (N
a l l ava i lab le ORNL swirl-flow heat ing da ta . /N2 ) Gr Re a ' Equation 2.10 adequately c o r r e l a t e s
I . I b .
Table 2.5. Forced-Convection Burnout Tests w i t h Ethylene Glycol
b .
e Test Section tbl, 'bo, (va)bo, (tsat)bo, (tb)bo' ('3tsub)bo' '/gl %o 3 -
O F OF "F geesd Lh lo-= B t u / h r - f t 2 - T e s t
No. Di, in. I , , i n . Material y" E, pin . nusb "F p i a C fps
1
2
3
4
5
6
7
8
9
10
11
l2
13
14
15
16
0.249
0.249
0.136
0.249
0.249
0.249
0.249
0.249
0.245
0.249
0.249
0.249
0.249
0.249
0.249
0.249
11.91
11.88
13.81
11.98
11.98
11.98
11.98 11.98
11.98
11.98
12.94
12.10
7.25
7-19
11.98
11.9
Copper 4.73
Copper 2.26
Copper 2.43
Copper 7.73
Copper 11.81
Copper 12.10
Copper 12-05
Copper 7.74
Copper 2.18
Copper No tape
Copper No tape
Copper No tape
347 ss 2.45
347 ss 2.38
Copper No tape
Copper No tape
17.. 7
24.3
63.3
43.3
27.9
20.4
27.5
21.5
22.7
32.9
29.6
37.8
27.3
28.2
25.8
f
138
72
148
73
127
98
115
73
145
124
127
149
111
103
124
122
41.0
279.0
140.0
170.0
43.9
20.4
33.3
19.9
27.1
19.5
31.7
31.4
36.9
64.2
87.7
29.1
25.4 427
72.0 536
63.2 493
38.0 506
62.3 431
35.0 398
97.5 418
28.2 397
66.8 408
17.3 3%
88.4 415
60.2 415
21.4 439
30.6 450
67.3 467
40.6 412
337 90
117 419
416 77
80 426
26 4 167
259 139
234 184
252 %
311 97
240 156
234 181
260 155
115 324
108 342
203 264
227 185
212
7510
9190
178
204
62
484
98
6940
1
1
1
564
1221
1
1
0.96 0.26
0.96
0.04
O.%
0.95
0.95
0.96
O . %
0.96
0.89
0.96
0.07
0.07
0.66
0.94
3.68
9.00
5.32
4.38
5.99
3.97
8.11
3.42
8.03
2.06
6.25
4.89
2.33
3.40
6.23
3.42
aChecked by x-ray measurements. bAverage value determined after t e s t a t 3 and 8 axial positions. ' A l l pressures are subcri t ical .
'Calculated from equation based on ro t a t ing slug-flow model (ref. 1).
fNot measured. local 4, a t the burnout s i te , which is the value listed, differed from the tube-average $bo by <5$ i n a l l cases.
40
4. With sa tura ted pool bo i l ing , a x i a l forced-convection l o c a l
boi l ing, and s w i r l forced-convection l o c a l bo i l ing , respect ively, e th-
ylene glycol i s character ized, f o r the var iab le ranges of t h i s study,
by burnout heat f luxes -l/3, 1/2, and 3/4 those f o r water a t t h e same
conditions of geometry, ve loc i ty , pressure, and bulk temperature.
2.4.2. Swirl-Flow Boiling with Net Vapor Generation
Test sect ions were designed and f ab r i ca t ed f o r use i n t h e two-phase
These t e s t sec t ions a re 1/4- and 1/2-in.-ID swirl-flow boi l ing s tud ie s .
A-nickel tubes 22 i n . long f i t t e d with i n t e r n a l twis ted tapes of O.Ol5-in.-
t h i ck Inconel. The i n i t i a l object ive of t he study i s the determination
of c r i t i c a l heat f luxes over the following var iab le ranges: e x i t pres-
sure = 15 t o 30 ps i a , i n l e t water ve loc i ty = 1 t o 3 fps, tape t w i s t r a t i o =
2.5 t o 8.0 ins ide diameters per 180-deg t w i s t , i n l e t subcooling = 0 t o
30°F, e x i t qua l i t y = 30 t o 80 w t 4, a l l with a v e r t i c a l o r i en ta t ion of
t h e t e s t sec t ion . Three heat exchangers - a preheater , a condenser, and
a l i q u i d re turn cooler - have a l s o been designed and ordered.
41
3. FLOW DYNAMICS AND TURBULEIVCE
*,
3.1. Vortex Fluid Mechanics
J. J. Keyes, Jr.
Consideration of the unique f l u i d mechanical propert ies of vortex-
type flows has s t i m l a h e d i n t e r e s t i n the appl icat ion of these flows t o
c e r t a i n advanced energy conversion concepts. The magnetohydrodynamic
vortex power generator, f o r example, u t i l i z e s a plasma i n vortex flow
in te rac t ing with a magnetic f i e l d t o generate an e l e c t r i c field."'
The cavi ty gaseous f i s s i o n reactor concept described by Kerrebrock and Meghreblian2' depends on a vortex ve loc i ty f i e l d f o r containment of the
f iss ionable mater ia l . These appl icat ions require operation with a high
r a t i o of tangent ia l t o r a d i a l through-flow ve loc i ty ; hence, energy losses
due t o viscous d iss ipa t ion through turbulence must be minimized s o t h a t
t h e energy required t o s u s t a i n the vortex w i l l be small compared with
the energy output of the device. \
Vortex-type flows are a l s o of general concern i n connection with
separation processes i n t h e nuclear f i e l d ; e .g . , gas-phase separation of
isotopic species, liquid-vapor separation i n chemical processing and i n
Rankine cycle space power systems, and l iqu id-so l id separation i n chemi-
c a l processing. There is a l s o much i n t e r e s t i n flows having a s t rong
tangent ia l component (e .g., s w i r l flow) f o r heat- t ransfer appl icat ions,
as discussed i n Section 2.4 of t h i s report .
flows a r e c l o s e l y re la ted , p a r t i c u l a r l y w i t h regard t o boundary-layer
phenomena. I n considering t h e use of vortex flow f o r gas separation o r
f o r heat t r a n s f e r , a knowledge of the i n t e r r e l a t i o n s h i p of the vortex
s t rength and the turbulent energy d iss ipa t ion is of fundamental importance.
I n t e r n a l s w i r l and vortex
"%. S. Lewellen and W. R . Grabowsky, ".Nuclear Space Power Systems Using Magnetohydrodynamic Vortices," Am. Rocket Soc . , 693-700 (May 1962).
"'J. L. Kerrebrock and R . V. Meghreblian, "An Analysis of Vortex Tubes f o r Combined Gas-Phase Fission-Heating and Separation of the F is - sionable Material, " U S A N Report ORNL-CF-57-11-3, Oak Ridge National Laboratory, Apri l 11, 1958.
42
3.1.1. . Gas Dynamics Studies
The vortex gas dynamics work a t ORNL has aimed a t providing t h e
following basic f l u i d mechanical information needed i n evaluat ing the
various appl ica t ions of je t -dr iven vortex f low:
1. Vortex s t r eng th (per ipheral , t angen t i a l ve loc i ty , or Mach
number) as a funct ion of such key var iab les as tube diameter, mass flow
r a t e , i n j ec t ion ve loc i ty and geometry, and pressure, with the a i m of
obtaining the maximum vortex s t r eng th pe r u n i t of input power.
2. Turbulence l eve l s as functions of appropriately defined Reynolds
moduli.
3. The separat ion c h a r a c t e r i s t i c s of vortex flow f o r a mixture of
l i g h t and heavy gases, including e f f e c t s of turbulence.
The experimental work conducted thus far, much of which is reported
i n Refs. 30 through 34, has provided a fa i r e luc ida t ion of item 1; and
it now appears possible t o generate vortexes i n small diameter tubes
(1.0 i n . or l e s s ) of s u f f i c i e n t s t rength t o be of preliminary i n t e r e s t
f o r gaseous reac tor and o ther appl'ications. Due t o t h e high l e v e l of
30J. L. Kerrebrock and J. J. Keyes, Jr., "A Preliminary Experimental Study of Vortex Tubes f o r Gas-Phase F iss ion Heating," UXAM: Report ORNL- 2660, Oak Ridge National Laboratory, February 6, 1959.
31J. J. Keyes, Jr. and R . E . Dial, "An Experimental Study of Vortex Flow for Application t o Gas-Phase F iss ion Heating, O a k Ridge National Laboratory, Apr i l 16, 1960.
.32J. J. Keyes, Jr., "An Experimental Study of G a s Dynamics i n High Velocity Vortex Flows," Proceedings of t h e 1960 Heat Transfer and Flu id Mechanics I n s t i t u t e , pp. 31-46, Stanford Universi ty Press, Stanford, Cal i forn ia , 1960.
i n Vortex Tubes with Application t o Gaseous F iss ion Heating," Am. Rocket e., 1204-1210 (September 1961).
F lu id Mechanics, S ta tus Report . July 1, 1959 - February 29, 1960," USAEC Report om-c~-60-10-6 , Oak Ridge National Laboratory, October 1960.
USAEC Report 0~~~-2837,
33J. J. Keyes, Jr., "An Experimental Study of Flow and Separation
3%. W. Hoff'mn e t a l . , "Fundamental Studies i n Heat T r m s f e r and
43
turbulent shear near the periphery, however, the e f f ic iency of vortex formation has been low i n the experiments c a r r i e d out thus far.'
preted i n t e r m of a p r a c t i c a l device, t h i s means t h a t bleed-off ( r a d i a l l y
outward. o r a x i a l l y ) of a s igni f icant f r a c t i o n of the i n l e t mass flow would
be required t o s u s t a i n the vortex without exceeding the allowable e x i t
mass' flow r a t e as determined by d i f fus iona l v e l o c i t i e s .
I n t e r -
Preliminary separat ion experiments have been ~ e r f o r m e d ~ l - ~ ~ which
suggest t h a t turbulence may be e f f e c t i v e l y suppressed by the strong den-
s i t y gradient i n the c e n t r a l region of the vortex f i e l d . The e f f e c t of
turbulence i n the outer region of the vortex on separation has been quan-
t i t a t i v e l y ascertained, but i n t u i t i v e l y it i s believed t h a t only low-tur-
bulence leve ls can be t o l e r a t e d i n t h i s region.
Specific vortex tube geometries s tudied i n the i n i t i a l phase of t h e
i n v e ~ t i g a t i o n ~ ~ - ~ ~ are described i n Table 3.1. depicting four t y p i c a l vortex tube geometries s tudied. For a l l geometries,
the r a d i a l s t a t i c pressure d is t r ibu t ion , from which tangent ia l v e l o c i t i e s
were determined by the method described i n Ref. 31, w a s measured by means
of pressure taps d r i l l e d i n the closed end of t h e tube.
Figure 3.1 is a photograph
The nomenclature employed i n cor re la t ing the data i s as follows:
M tangent ia l Mach number a t the tube periphery
M i n j e c t i o n Mach number P
j
m = 2 7 ~ r Q p mass flow r a t e , . l b / s e c - f t
r a d i a l Reynolds modulus
- - - %!-!dh t angent ia l per ipheral Reynolds modulus t , P pP
NRe
M (r' = 0 . 8 ) a =
j M
j e t ve loc i ty recovery r a t i o
§Efficiency of vortex formation i s d i r e c t l y proportional t o vortex s t rength and inversely proportional t o t h e ne t power input required t o sus ta in the. vortex.
4 4 z
I
2-in.- DlAM POROUS TUBE WITH !-in.- AND 0.6- in.- DlAM INSERTS
(-in - DIAM TUBE SLOTTED FOR BOUNDARY - LAYER BLEED OFF
2-in.- DlAM SLIT-FED TUBE
ASSEMBLY OF (-in - DIAM LAMINATED TUBE; WIDE WASHERS CORRUGATED TO GIVE A TOTAL OF 360,000 IN- JECTION JETS
Fig. 3.1. Typical Vortex Tubes.
.
45
Note t h a t t h e only dimensional quant i ty i s m, t h e mass flow r a t e ; Qr and
&p are the r a d i a l and t angen t i a l per iphera l veloci ty , respect ively; p and
p are f l u i d densi ty and absolute v i scos i ty , respec t ive ly .
Table 3 .l. Description of Experimental Vortex Tubes
In j ec t ion Geometry Exi t Radius Tube Tube D i m or Radial
Width, Pos i t ion Tube Radius Tube ID, rd> Len@’h, Descr ip t ion I i n . r r a e No. i n . i n .
1 2.00 12.0
2 2.00 12.0
2.00 12 .0 3 7
3A 1.00 12.0
3B 0.63 3 -25 , 6.0,
12.0
4A 2 .oo 12.0
12 nozzles 0.0135 0.92 0.140
12 nozzles 0.0100 0.84 0.140
12 nozzles 0.0135, 0.92 0.0625 -0.250 0.0330
12-164 0.020 0.90 0.280-0.500 nozzles
11-54 0.020 0.90, 0.380-0.525 nozzles 0.95
12-in.-long 0.002 1.00 0.140 s l i t
4B 2 .oo 12.0 12-in.-long 0.002 0.92 0.140 s l i t
5 0.64 3.5 12 nozzles 0.028 0.88 0.308-0.362
6 0.315 1.75 60 nozzles 0.0135 0.90 0.3-0.5
The values of per iphera l t angen t i a l Mach number, M which were
observed i n 0.64- and 2.0-in.-diameter tubes with N, gas in jec ted a t room
temperature and a t sonic ve loc i ty , M . = 1, are indicated i n Fig. 3.2 as a J
function of the mass flow r a t e . The da ta include measurements with tube
w a l l pressures of 83 and 108 ps i a . It must be pointed out t h a t , based
on the o r ig ina l laminar flow a n a l y ~ i s , ~ ’ a per iphera l Mach number ap-
proaching un i ty should be achievable a t the allowable mass flow r a t e s of
P’
0.01 t o 0.02 l b / s e c . f t . Note, however, t h a t i n t h i s range of mass flows
r?
M
p,
TA
NG
EN
TIA
L
U. 00
9 .
PE
RIP
HE
RA
L
MA
CH
. N
O.
0
0
dw
0 000
. .
.
L
N
ID
.
a
0 s
, ,
e
47
.
the values of M
f o r the 2-in. tube, and from approximately 0.2 t o 0 .3 f o r the 0.64-in. tube.
in.-diameter tube, and required a mass flow r a t e of 0.06 lb / sec . f t .
t he observed ve loc i t i e s were s ign i f i can t ly lower than would be expected
f o r laminar flow, it i s probable that boundary layer turbulence ex is ted
with correspondingly high w a l l shear .
ac tua l ly observed var ied from approximately 0.10 t o 0.18 P
The highest per ipheral Mach number of 0.49 w a s observed i n a 0.64-
Since
Figure 3.3 i l l u s t r a t e s the e f f e c t of decreasing tube diameter on the
tangent ia l Mach number t o inject . ion Mach number r a t i o p lo t t ed as a func-
t i o n of r a d i a l pos i t ion . The humps i n t h e curves f o r t h e 0.64- and 0.20-
i n . tubes may be associated with secondary flow and/or in jec t ion e f f e c t s .
Note t h a t , as t h e tube diameter decreases with corresponding decrease i n
, the ve loc i ty r a t i o general ly increases . The concave downward
shape of the p r o f i l e f o r the 0.32-in.-diameter tube suggests t h a t t he
turbulent boundary layer may, i n t h i s case, represent a s igni f icant f r ac -
t i o n of t he tube radius . Three-dimensional flow e f f e c t s a l so p lay an
important ro le .
NRet ,p
Preliminary estimates of the degree of turbulence i n vortex flow
were made by comparison of observed ve loc i ty p r o f i l e s with published
solut ions of the Navier-Stokes equations i n terms of v i r t u a l v i scos i ty
(molecular v i scos i ty plus eddy v i scos i ty ) .
r a t i o p*/p of v i r t u a l t o molecular v i scos i ty near t he periphery as a
function of tangent ia l per ipheral Reynolds modulus,
tubes s izes , N2 and He gas.
p*/p z 30; and a t t h e highest Reynolds modulus , 1 . 6 x lo6, the r a t i o i s
near ly 700. The r a t i o would, of course, be un i ty f o r laminar f low.
Since the operating Reynolds modulus may be i n the range from 1 X lo5 t o
1 x lo6, it i s suggested t h a t turbulence w i l l probably e x i s t near t he
periphery.
Figure 3.4 shows the observed
f o r th ree
A t t he lowest Reynolds modulus, 4 x l o 4 ,
It must be pointed out t h a t t he estimates of v i r t u a l v i scos i ty a re
based on a two-dimensional flow analysis i n which the r ad ia l mass flow
i s assumed uniform and equal t o t h a t which exhausts a t the center of the
1.3
1.2
1 . 1
1.0
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0. I
r', RADIAL POSITION
Fig. 3.3. Comparison of Ta.ngentia1 Mach Nwnber P r o f i l e s f o r J e t - Driven Vortex Tubes of 0.32-, 0.64-, and 2.0-in. Diameter.
P
. 49
1000
5 0 0
200
100
5 0
20
10 . -
1 2 5 10 20 50 1 0 0 x 1 0 ~ 2.5 TIMES PERIPHERAL TANGENTIAL REYNOLDS NUMBER, 2.5 N R ~ , , ~
Fig. 3.4 . Variation of Ratio of Virtual Viscosity to Molecular Vis- cosity with Peripheral Reynolds Modulus. 0.6-, 1.0-, and 2.0-in.-diam tubes; N2 and He gas.
.
tube.
ca t e t h a t t h i s i s not r e a l l y t r u e and t h a t same of t he flow i s ca r r i ed
i n t h e boundary layers on the w a l l and end p l a t e s . Neglect of t h i s
component of t he flow means t h a t t he v i s c o s i t i e s estimated i n t h i s work
a re upper limits.
Measurements by Kenda1135 a t the J e t Propulsion Laboratory ind i -
The i n t e n s i t y of turbulent w a l l shear i s of obvious i n t e r e s t i n
r e l a t i o n t o f l u i d friction.characteristics of vortex tubes . Estimates
of maximum coe f f i c i en t s of sk in f r i c t i o n , C have been made from tangen-
t i a l ve loc i ty and v i r t u a l v i scos i ty data, making use of a torque balance
i n which the dr iving torque due t o the decrease i n momentum of the feed
j e t s i s equated t o the sum of the torque due t o w a l l shear and t h e i n t e r -
na l f l u i d torque .32 obtained i n t h i s manner
versus the per iphera l t angen t i a l Reynolds modulus, including da ta f o r
5/8-, 1-, and 2-in.-diameter vortex tubes.
least-squares l i n e represented by the equation:
f '
Figure 3.5 i s a graph of C f
The data s c a t t e r about the ,
-0 * 27
The lower (dashed) curve i s a p l o t of t he Schoenherr average l a w f o r
turbulent C on a f l a t p l a t e w i t h zero pressure gradient . Since the
Schoenherr equation i s based on a length Reynolds modulus, which i s not
d i r e c t l y comparable with the Reynolds modulus as defined f o r vortex
flow, the 1,ower curve should be considered f o r reference purposes only.
f
Separation experiments were c a r r i e d out u t i l i z i n g helium and a
heavy fluorocarbon vapor C&FI6 (molecular weight 400). The heavy gas
w a s introduced as a d i l u t e mixture with helium through t h e porous w a l l
of the 2-in.-diameter vortex tube described i n Fig. 3.1. probe introduced r a d i a l l y a t the midaxial pos i t i on w a s used t o withdraw
gas f o r ana lys i s by measurement of thermal conduct ivi ty .
An 0.008-in.-OD
A re la t ionship between the t angen t i a l Mach number at t h e radius of
m a x i m u m mole f r a c t i o n of heavy component and t h e e x i t mass flow r a t e i s
35James M. Kendall, Jr., "Experimental Study of a Compressible Viscous Vortex," Cal i forn ia I n s t i t u t e of Technology Technical Report No. 32-290, J e t Propulsion Laboratory, June 5, 1962.
.
51
0.1
0.05
I- z w 0
0.02 w ou z 0 F 0 0.01 E z Y v)
W 0
-
2 0.005 a W > c e
0.002
0.001
UNCLASSIFIED ORNL-LR-DWG 48875
e 2-in. DIA TUBE CI l - i n . DIA TUBE
GRADIENT, TURBULENT FLOW
logfo (‘f NRe, ) (SCHOENHERR)
= 0.242 Cf-Oe5
8 10 20 50 loo 200 ( ~ 1 0 ~ )
PERIPHERAL TANGENTIAL REYNOLDS NUMBER NRet, p ’
LENGTH REYNOLDS NUMBER ( P L A T E ) NRem 9
Fig. 3.5. Comparison of Skin Friction Coefficients for Vortex Tube and Flat Plate.
52
derived i n Ref. 29 f o r laminar flow. For the spec ia l case of a very d i l u t e mixture, as employed i n the experiments, the re la t ionship i s
where M i s the t angen t i a l Mach number a t the radius of maximum mole
f r ac t ion ; m
r a t i o of mole f r ac t ions of heavy component i n the e x i t gas stream and
a t t h e peak; ( p D12)m, t he density-molecular d i f f u s i v i t y product a t t h e
peak; m 2 / q , the mass r a t i o ; and y = c /e . equation, it i s seen t h a t M va r i e s d i r e c t l y as t h e square root of the
e x i t mass flow r a t e and inversely as the square root of t he d i f f u s i v i t y .
I n view of t h e small magnitude of d i f f u s i v i t y f o r C g 1 6 i n helium near
room temperature, t h e equation can be s a t i s f i e d f o r values of M which
a re experimentally a t t a inab le only i f t h e e x i t mass flow r a t e i s low.
Thus, it w a s necessary i n the experiment t o introduce excess mass flow
i n order t o generate the required vortex s t rength and t o bleed of f a l l
but t he allowable e x i t flow.
m t he e x i t pass flow r a t e per u n i t tube length; xe/x e ' m, the
From an examination of t h i s P V
m
m
Operating conditions and r e s u l t s f o r th ree t y p i c a l separat ion runs
a re s m r i z e d i n Table 3.2. The peaks i n mole f r a c t i o n of heavy com-
ponent were observed t o occur a t pos i t ions varying from 11 t o 21% of the
radius . The t angen t i a l Mach numbers observed a t t he radius of peak mole
f r ac t ion were determined w i t h t h e sample probe in se r t ed and, thus, r e -
f l e c t t he e f f e c t of t he probe on the vortex s t rength . These Mach numbers
were 0.70, 0.70, and 0.55, respect ively.
Mole f r a c t i o n r a t i o s xe/x of 0.5 and 0.67 were measured f o r runs 1 m and 2. The ca lcu la ted Mach numbers a t the peak pos i t i on f o r runs 1 and
2 were obtained from the preceding equation, using an experimental value
of molecular d i f f u s i v i t y f o r C g 1 6 - & , corrected f o r temperature. Note
t h a t the observed and ca lcu la ted values of M agree t o within 26. m
From these i n i t i a l exploratory experiments, it w a s concluded t h a t
high w a l l f r i c t i o n due t o turbulence i n t h e boundary l aye r on t h e inside
surface of t he vortex tube is an important f a c t o r l imi t ing t h e attainment
53
of high vortex s t r eng th a t low mass throughput.
techniques f o r possible reduction i n boundary layer turbulence w a s t he re -
fore i n i t i a t e d . For example, experiments were ca r r i ed out with uniform
w a l l bleed-off, u t i l i z i n g 2-in.-diameter nozzle and s l i t - f e d tubes 3 and
4 with porous w a l l s , f o r the puspose of determining whether reduction
i n w a l l shear by boundary-layer s t a b i l i z a t i o n could be observed.
case, however, w a s a s ign i f i can t increase i n vortex s t rength measured
f o r bleed flows up t o three times the e x i t flow. The vortex s t rength
was observed t o increase with extreme w a l l cooling, however, suggesting
t h a t s t rong dens i ty and v i scos i ty gradients i n the boundary l aye r may
reduce turbulent shear . Additional study of t h e influence of tempera-
t u r e gradients i s planned.
A n inves t iga t ion of
In no
Table 3.2. Summary of Operating Conditions and Results f o r He-Cgls Separation Experiments i n a 2-in.-diameter Vortex Tube
Run Parameter 1 2 3
I n l e t helium mass flow ra t e , m. ( l b / s e c . f t ) 0.023 1
Bleed r a t i o , % 15.6 ( P s i 4 56.3 ' pP
Wall pressure
Exi t diameter, 2re ( i n . ) 0.221
Bleed off pos i t ion , r' 0.25, 0.40
0.15
Observed Mach number a t r&, bL 0.70 Observed ra;io of e x i t mole f r ac t ion t o 0.50
Calculated M, 0.58
Ratio of observed % t o ca lcu la ted
I Observed mole f r a c t i o n peak pos i t ion , rm
t h a t a t rm, xe/xm
1 . 2 ,
0.030
9.4
0.250
40.8
1.00 ( w a l l s l i t ) 0.11 0.70
0.67
0.81
0.87
0.030
17.5 53.0 0.250 0.40
(axial)
0.21
0.55 -
- -
Measurements were a l s o made of ve loc i ty d i s t r ibu t ions i n a 1 - in . -
diameter by @in.-long tube, i n which the boundary l aye r could be bled
of f throdgh c a r e f u l l y shaped and or iented w a l l s l i t s (Fig. 3.1, lower
l e f t ) . The gas was icjec-ted through 60 small nozzles ( three rows of
54
20 each), and bled-off through O.OO5-in. w a l l s l i t s spaced about 0.1 i n .
apar t c i rcumferent ia l ly . It w a s o r i g i n a l l y thought t h a t a reduction i n
boundary-layer turbulence might be achieved i n much the same fashion as
as f o r t he case of a s l o t t e d a i r c r a f t wing w i t h suc t ion .
e f f e c t on w a l l f r i c t i o n w a s observed, Eut t he model d id a f fo rd a means
of inves t iga t ing a technique f o r uniform removal of t h e excess mass flow
required f o r generation of' s t rong vortexes.
No s ign i f i can t
Figure 3.6 is a t y p i c a l t angen t i a l Mach number p r o f i l e obtained with
s l i t bleed-off. Note t h a t 94% of t h e i n l e t mass flow w a s removed through
the s l i t t e d w a l l , w i t h 6% (or 0.024 l b / s e c . f t ) exhausted through t h e e x i t
o r i f i c e loca ted a t one end of t he tube. With an in j ec t ion Mach number of
about 1 . 3 and 50 p s i a tube pressure, a per iphera l Mach number of 0.64 w a s
generated with a maximum i n the ve loc i ty of M = 1.05.
Fina l ly , some experiments have been c a r r i e d out i n which vortexes
were generated by flow through a uniformly porous w a l l , t h e pores of which
were or ien ted near ly t angen t i a l ly (Fig. 3.1, lower r i g h t ) . By g rea t ly
increasing the number of j e t s ( i n t h i s case pores) and thus decreasing
the spacing between adjacent j e t s , it w a s reasoned that, i n t h e l i m i t ,
a s i t u a t i o n within t h e boundary l aye r would be approached i n which adja-
cent j e t s would r ide on each other , thus forming a ro t a t ing gaseous " w a l l "
with minimum in t e rac t ion a t the s o l i d surface.
the very small equivalent diameter of each individual pore, t h e j e t flow
wo-J ld l i k e l y be lamir,ar; and, hoperully, t he re would r e s u l t l e s s turbu-
lence i n the boundary l aye r . It is apparent t h a t t o a t t a i n t h i s r e s u l t
would indeed require an enormom number of microscopic pores, s ince the
allowable void f r a c t i o c of t he w a l l is low.
Furthermore, because of
A n i n i t i a l e f f o r t t o develop a mater ia l with t angen t i a l ly or ien ted
pores w a s made. The f i rs t tube u t i l i z i n g t h i s method of vortex generation
consis ted of a s t ack of 2-mil brass washers, a l t e r n a t e washers having been
corrugated by means of a die-stamp operat ion described i n Ref. 36. e r s (0.2-in. wide) stamped with t h e equivalent of 120 corrugations (cor-
responding t o 240 pores pe r washer) were stacked a l t e r n a t e l y w i t h O.05-in.
Wash-
. .
.
~
'%. J. King and E. G. Fa r r i e r , "Construction of Oriented Flow Devices, December 7, 1962.
U S E Report. K-1549, Oak Ridge Gaseoils Diffusion Plant ,
55
Fig. 3.6. Typical Tangential Mach Number P r o f i l e for a Slitted Vor- tex Tube w i t h Wall Bleed-Off.
plane spacer washers t o give s l i t - l i k e pores about 1 4 m i l s X 0.1 m i l i n
c ross sec t ion and 65 m i l s i n length. In j ec t ion Reynolds moduli f o r pores
cf t h i s s i z e are i n the range from 10 t o 100. Figure 3.7 i l l u s t r a t e s t he
f i n a l washer assembly, which w a s 1 - in . I D x 6 i n . i n length; t he re were
3000 washers with 360,000 pores i n t h i s device.
were employed t o vary the pressure on the s tack, and thus t o permit wide
va r i a t ions i n poros i ty . Gas entered from the annular jacket and was re -
movedthrough a 0.5-in. o r i f i c e a t one end. Taps f o r measuring the radial
pressure gradient were provided a t both ends of the tube .
Four adjustable t i e b o l t s
Figure 3.8 compares values of t h e j e t ve loc i ty recovery r a t i o ,
a! = M o . 8 / ~ j , observed with the multipored laminated tube ( s o l i d c i r c l e s )
with values observed using conventional j e t -driven tubes of O . p , 0.64-, and 1.0- in . diameter. For laminar vortexes, a! should vary with radial
Reynolds modulus and should be a t l e a s t 0 .9 f o r N
f r o m t h i s f igure t h a t , when compared on t h e bas i s of NRe,r, t h e values of
a! f o r the laminated tube f a l l general ly above those f o r conventional j e t -
driven tubes; and, indeed, i n some runs at high N (Aoo), values ap-
proaching uni ty were observed f o r a!. The f a c t t h a t a! may be l e s s than
t h a t predicted by laminar theory a t N
at NRe, r t he flow f i e l d near t h e w a l l is probably tu rbu len t . Additional data a t
low N ~ e , r
L 1 00 .~ ~ It is seen Re,r
Re,r
<800 (as evidenced by the point Re,r = 500) suggest t h a t , a s with the conventional je t -dr iven tubes,
i s obviously -needed t o support t h i s observation.
The comparison i n terms of r a d i a l Reynolds modulus fa i l s t o take i n t o
account t he f a c t t h a t i n j ec t ion v e l o c i t i e s f o r t h e laminated tube were .
from 2 t o 10 times lower than f o r t he "conventional" vortex tubes, a! w a s
found i n previous s tud ies t o depend on t h e in j ec t ion velocity.31
attempt t o take t h i s i n t o account i s shown i n Fig. 3.9, i n which a is
p l o t t e d against t h e r a t i o N
t he data f o r the laminated tube ( s o l i d c i r c l e s ) with data fo r the other
tubes, b e t t e r than does the p l o t of a! versus NRe,r, t h i s apparent improve-
ment might be due t o t h e f a c t t h a t M . appears i n the denominator i n both J
ordinate and abscissa , thereby producing perhaps a b e t t e r co r re l a t ion
An
/M . Re,r j While t h i s p l o t appears t o co r re l a t e
..
37M. L. Rosenzweig, "Sumnary of Research i n the F i e l d of Advanced Nuclear Propulsion - January-June 1961, " Report TDR-594 (1203-01)STR, Space Technology Laboratory.
UNCLASSIFIED O m - L R - D W G . 67928
18 RADIAL PRESSURE II RADIAL PRESSURE TAPS (0.0135"I.D.) TAPS (0.0135" I.D.)
I500 ea. CORRUGATED WASHERS (0.002" x 0.20")
0 0 e a . SMOOTH WASHERS (0.002"~ 0.05")
ADJUSTING SCREWS
SECTION A - A
Fig. 3.7. Multipored Laminated Vortex Tube Assembly.
58
cn 0 00 0
k
59
--
<
ORNL-LR
- 00-0- * 0
-0
x
--LEGEND -
.
DWG.
--
0.2 0
6835
-
El - 0.32 in. dia. A - 0.64 in. dia. X - 1.0 in. dio. 0 - 1.0 in. dia.
--
LAMINATED (3.6 X 10' JETS)--
--
RADIAL REYNOLDS NUMBER JET MACH NUMBER
Fig. 3.9. Variation in Jet Velocity Recovery Ratio with Radial Reynolds Modulus Divided by Jet Mach Nurriber f o r Conventional and Multi- pored Laminated Vortex Tubes.
60
than the data warrant. If t h i s p l o t does co r re l a t e s i g n i f i c a n t l y the
and M then it would appear t h a t recove,ry r a t i o s e f f e c t of both NRe,r
f o r t he multipored, laminated vortex tube a re somewhat higher than those
f o r tfconventionaltf 1 - in . -diameter je t -dr iven vortex tubes (crosses) , but
not s i g n i f i c a n t l y higher than t h e r a t i o s observed i n 0.32-in. -diameter
tubes (open squares) ; t h i s ind ica tes t h a t many times the number of pores
used i n t h i s model, and hence much smaller pores, may be necessary t o
a t t a i n the l imi t ing condi t ion of a " ro t a t ing w a l l " of high ve loc i ty
gaseous j e t s . Techniques f o r increasing the number of pores without
increasing the f r e e a rea or pore length-to-diameter r a t i o a r e under
study. 36
j '
3.1.2. Magnetohydrodynamic Studies
J. J. Keyes, Jr. W . K . Sar tory T . S. Chang
A s has been indicated, attempts t o reduce turbulence l e v e l s i n vortex
flows by purely hydrodynamic means have met with l imi ted success. An a l t e rna t ive method which can, i n pr inc ip le , produce a s i g n i f i c a n t stabi-
l i z i n g influence over the e n t i r e flow f i e l d involves t h e in t e rac t ion
between the ve loc i ty f i e l d i n an e l e c t r i c a l l y conducting medium and a
su i t ab ly or iented magnetic f i e l d , making use of t h e Lorentz force which
might be considered as a r e s to r ing body force exerted on a f l u i d volume
element when it i s displaced from i t s pos i t i on of equilibrium. Since i n the cases of power conversion and gaseous reac tor appl ica t ions operation
w i l l be a t s u f f i c i e n t l y high temperature t o sus t a in ion iza t ion i n the
gas (ei'ther thermally o r by alkal i -metal ion seeding), t he p o s s i b i l i t y
of magnetohydrodynamic (MHD) s t a b i l i z a t i o n should not be overlooked.
For example, Chang3' has shown t h a t a uniform axial magnetic f i e l d is
i n pr inc ip le capable of completely s t a b i l i z i n g an inv isc id , p e r f e c t l y
conducting, pure vortex flow (l/r ve loc i ty p r o f i l e ) aga ins t small per-
tu rba t ions without a l t e r i n g the mean flow. In t h e case of je t -dr iven
3&r. S. Chang, 'fMagnetohydrodynamic S t a b i l i t y of Vortex Flow - A Nondissipative, Incompressible Analysis," USAEC Report ORNL-TM-402, Oak Ridge National Laboratory, October 1962.
61
vortexes, however, t he j e t s themselves can introduce forced boundary
o s c i l l a t i o n which may prevent complete s t a b i l i z a t i o n , although p a r t i a l
s t a b i l i z a t i o n may be e f fec ted . I
A general ana lys i s of MHD e f f ec t s i n vortex flows3' has indicated
t h a t there e x i s t severa l promising appl ica t ions . One po ten t i a l advantage
of MHD is the p o s s i b i l i t y of u t i l i z i n g magnetic i n t e rac t ion e f f e c t s on
the end w a l l s which introduce forces ac t ing t o r e t a r d t h e shor t -c i rcu i t ing
inward r a d i a l boundary-layer flow discussed i n Refs. 35 and 40.
appl ica t ion is considered i n Ref. 41. This
Analysis of magnetic s t a b i l i z a t i o n of vortex motion based on a d i s -
s ipa t ive assumption (viscous f l u i d w i t h f i n i t e e l e c t r i c a l conduct ivi ty)
has a l s o been considered, and a c l a s s of exact s t a t iona ry solut ions of
incompressible, magnetohydrodynaniic flow w i t h c y l i n d r i c a l symmetry under
the act ion of an ex te rna l ly applied magnetic f i e l d i n the d i r ec t ion of
t he axis of symmetry have been obtained. These so lu t ions do not depend
on the f i n i t e conduct ivi ty of the f l u i d medium or on t h e ex terna l ly ap-
p l ied , uniform magnetic f i e l d .
Dimensionless forms of t he per t inent equations character iz ing the
s t a b i l i z a t i o n of pure, d i s s ipa t ive vortex flow have a l s o been obtained.
S imi la r i ty parameters include : /
, (3.3) § - ro Qo the per iphera l t angen t i a l Reynolds modulus, NRee - P
the per iphera l magnetic Reynolds modulus, RM = p,, (I Qo ro ,, (3.4)
39J. J. Keyes, Jr., "Some Applications of Magnetohydrodynamics i n Confined Vortex Flows, It USAEC Report ORNL-TM-479, Oak Ridge National Laboratory, February 28, 1963.
40F. N . Peebles and H . J. Garber, "Studies of t h e Swirling Motion of Water Within a Spherical Vessel, It USAEC Report om-c~-56-1-161, Oak Ridge National Laboratory, January 1956.
Vortex Flow over a Disk with an Axial Magnetic F ie ld ," Report No. ATN-63(9227)-1, Aerospace Corporation, December 7, 1962.
redefined i n terms of radius , so t h a t NRe
41W. S. Lewellen and W . S. King, "The Boundary Layer of a Conducting
'Note t h a t t he t angen t i a l per iphera l Reynolds modulus has been = 1/2 NRet w i t h Qo = &p;
p i s the molecular v i scos i ty . e ,P
I
62
the per ipheral Alfv&n modulus, A =.,/= Qo/Bo , (3.5)
(3.6) 2 t h e magnetic i n t e rac t ion parameter, S = CI Bo ro/p &o ,
and t h e per iphera l Hartmann modulus, H = ,,/- Bo r 0 . (3.7)
is t h e magnetic permeabili ty; u, t h e e l e c t r i c a l , Po I n these expressiocs
conductivity; and Bo, t h e appl ied magnetic f i e l d s t r eng th .
? The general formulation f o r t he magnetic s t a b i l i z a t i o n of d i s s ipa t ive
vortex flow w a s s impl i f ied f o r t h e case of marginal s t a b i l i t y w i t h very
la rge Reynolds modulus flow. It w a s found t h a t when t h e assumptions
N R ~ ~ >> 1, S 5 1 are s a t i s f i e d , t he c r i t e r i o n f o r marginal s t a b i l i z a t i o n
becomes equivalent t o that for nondissipative vortex
general c h a r a c t e r i s t i c so lu t ion f o r t h e magnetic s t a b i l i z a t i o n of d i s -
s ipa t ive ,vo r t ex flow is being considered, and an attempt w i l l be made t o
include t h e boundary-layer region i n t h e so lu t ion .
A more
An exploratory experimental study w a s i n i t i a t e d , which i s intended
t o e s t a b l i s h whether an MHD s t a b i l i z a t i o n e f f e c t does indeed e x i s t f o r
je t -dr iven vortex flow of a conducting f l u i d and t o determine t h e nature
and magnitude of t h i s e f f e c t i n s t a b i l i z i n g t h e flow aga ins t tu rbulen t
breakdown and i n inf luencing the secondary Plow s t ruc tu re . The apparatus
is diagrammed i n Fig. 3.10. The magnet is a 6.4-in.-ID a i r core solenoid
(water-cooled copper) located i n t h e OR% Thermonuclear Division Magnet
Laboratory.
0.246 var i a t ion r a d i a l l y and 104'0 v a r i a t i o n a x i a l l y f o r t he 1 .1- in . I D by
6- in . long vortex tube model used i n the experiments. A concentrated
aqueous so lu t ion of N H 4 C l (350 g / l i t e r ) is employed as t h e conducting
f l u i d .
operating temperature of 210"F, which is about t h a t of a low-temperature
alkali metal "seeded" plasma.
t o handle; i t s physical p roper t ies a re c lose t o those of water a t cor-
responding temperatures, and it permits t he use of flow v i sua l i za t ion
techniques. Figure 3.11 is an ove ra l l view of t h e apparatus, showing
the 62 kilogauss solenoid a t t h e l e f t .
tube i n s t a l l e d i n t h e solenoid.
This magnet provides a maxirmun f i e l d of 62.3 kilogauss, with
The conduct ivi ty of t h i s e l e c t r o l y t e i s 0.88 mho/cm at t h e maximum
Aqueous JJH4Cl is a r e l a t i v e l y easy f l u i d
Figure 3.12 depic t s t h e vortex
1 :.
. . . , .
' 7 . . - .. - -
. I . r
UNCLASSIFIED 0 R NL-LR-D WG. 78850R
10 MICRON TFILTER
40 psi (max 5 GPM
I SAT. Aq. N H 4 C L SOLN. (75-21OOF)
-8=
i! DRC
8"- l 5 psi
3Y-PASS i
Fig . 3.10. Schematic Flow Diagram for MHD Experiment No. 1.
64
65
Y .
66
The vortex flow f i e l d is generated by t angen t i a l i n j ec t ion through
two 0.012-in. s l i t s extending the length of t h e tube and f ed from an
annulus as indicated i n Fig. 3.13. hole a t the center of one end of t h e tube . The assembly i s fabr ica ted
of type 316 s t a i n l e s s s t e e l (nonmagnetic), with Plexiglas end w a l l s f o r
viewing i n the axial d i r ec t ion .
of dye i n t o the c y l i n d r i c a l w a l l boundary l aye r .
f i laments a t various pos i t ions (emerging from,direct ly downstream, half
way between, and d i r e c t l y upstream of a s l i t ) is the primary technique
f o r determining e f f e c t s of t h e magnetic f i e l d on flow s t a b i l i t y .
i n j ec t ion is a l so used f o r approximate determination of tangent ia l veloc-
i t y , when t h i s ve loc i ty is s u f f i c i e n t l y low (a0 cm/sec) f o r observation
of t h e time required f o r a puff of dye t o complete a prescr ibed number
of revolutions at a known radius . Ef fec ts i n the end w a l l boundary layer
a re made evident by observation of dye introduced through t h e end w a l l
opposite t he o u t l e t , .as indicated at the r igh t of Fig. 3.13. The dye
employed is a very d i l u t e so lu t ion of red coa l - ta r co lor i n sa tura ted
aqueous NH&1 t o provide a f l u i d having proper t ies e s s e n t i a l l y iden t i ca l
with those of the working f l u i d .
The so lu t ion e x i t s through a 1/8- in .
Also indicated a re t aps f o r i n j ec t ion
Observation o f t he dye
Dye
The i n i t i a l experiments have included:
1. Visual observation and photograph of dye pa t te rns i n the boundary
layers and i n the i n t e r i o r of the flow f i e l d , as influenced by the appro-
p r i a t e Reynolds moduli and by the magnetic parameters (Eqs . 3.4-3.7); emphasis has been on determination of t h e e f f e c t of t he magnetic f i e l d
on laminar flow s t a b i l i t y .
2. Determination of the influence of t h e magnetic i n t e rac t ion on
the t angen t i a l ve loc i ty near t he vortex tube periphery.
operating ternper.ature (from 77°F t o 210°F, producing s ign i f i can t var ia -
t i o n s i n e l e c t r i c a l conductivity, 6, and molecular v i scos i ty , p), mass
flow r a t e , m, and appl ied magnetic f i e l d s t rength, Bo, enabled operation
i n the range 4 < NRe
Variation i n
< 50, 200 < NRe < 6000, 0 < S < 4, and 0 < H < 41.
The most s t r i k i n g observation of the e f f e c t of t he magnetic f i e l d is
the supression of boundary-layer i n s t a b i l i t i e s which a re i n i t i a t e d a t
very low N R ~ .
r e
For example, with no magnetic f i e l d , t h e boundary l aye r e
.
,
67
UNCLASSIFIED ORNL-LR-DWG. 78852R
2 AND 4 ARE IN CENTER OF TUBE (3-1/8" FROM EACH END)
1 I S 1/2" ABOVE CENTER 3 I S 1/2" BELOW CENTER
MEASUREMENT
Fig. 3.13. 1 .1- in . -Dim Tangential Sli t-Fed Vortex Tube for MHD Studies.
68
on the concave cy l ind r i ca l w a l l exh ib i t s an osc i l la tory- type i n s t a b i l i t y
a t 260 < NRee < 410.
decelerat ion of t h e impinging j e t by t h e ac t ion of w a l l shear; t h e decele-
r a t ed f l u i d "separates" momentarily from t he w a l l and i s convected rad i -
a l l y by the ac t ion of t h e r a d i a l pressure gradient induced by t h e mean
t angen t i a l motion. The phenomenon thus has c h a r a c t e r i s t i c s similar t o
This i n s t a b i l i t y may a r i s e as a r e s u l t of the rap id
those of t he Taylor-Goertler i n s t a b i l i t y f o r flow over concave w a l l s .
The j e t i n t e rac t ion complicates t h e mechanism, however. The large-scale
per turbat ions i n flow near t he w a l l appear t o t r i g g e r a general d i s -
ordering of the motion i n the i n t e r i o r of t he f l u i d , r e su l t i ng i n gross
mixing and disappearance of any well-defined laminar p a t t e r n s .
3.14 includes photographs of dye t r a c e s in j ec t ed a t the tube w a l l 1 /4 i n .
downstream from a s l i t (mid-axial pos i t i on 3, Fig. 3.13, l e&) . '
t i o n i s clockwise i n the f i g u r e . For t he photograph a t the l e f t , there
is no appl ied magnetic f i e l d , NRe = TOO (H = S = 0) , and the boundary-
l aye r i n s t a b i l i t y j u s t discussed occurs. The dye fi lament becomes un-
s t ab le i n about 1/4 revolut ion and gross r a d i a l mixing i s evident, p a r t i -
c u l a r l y near the periphery. A s depicted i n the photo a t the r i g h t of
Fig. 3.14, appl ica t ion of a 62 kilogauss a x i a l magnetic f i e l d with
NRe, = (H = 39, S = 1.8) completely suppresses the boundary-layer i n s t a -
b i l i t y and t h e e n t i r e flow f i e l d exh ib i t s a general ly well-defined laminar
c h a r a c t e r i s t i c . Although t h e dye fi lament remains i n t a c t , s m a l l per tur -
bat ions which are bel ieved t o be induced by vortex shedding from the l i p s
of the in j ec t ion s l i t s p e r s i s t for severa l revolut ions.
i n t e rac t ion e f f e c t depends on t h e c h a r a c t e r i s t i c length or sca le of t h e
disturbance, it is not surpr i s ing t h a t t h e f i e l d fa i l s t o suppress these
very small eddies. The f i e l d does function, however, t o prevent growth
of the small eddies in to f u l l - s c a l e dis turbances.
Figure
Rota-
e
Since t h e magnetic
The 62-kilogauss f i e l d w a s observed t o completely s t a b i l i z e the
boundary flow up t o N R ~
Above t h i s value of t angen t i a l Reynolds modulus , i n s t a b i l i t i e s occur
near t he periphery but t he amplitude of t h e assoc ia ted eddies and the
of about 1600 at 210'F (H = 41, S = 1.08). e
The dye t r a c e s were photographed using 35 mm Eastman Kodak Company Y
Ektachrome film with an exposure time of l/l7OO sec .
69
amount of r a d i a l mixing is reduced as compared t o t h e case of no magnetic
f i e l d . is
above the t r a n s i t i o n value of 1200 f o r t h i s p a r t i c u l a r run a t 200°F,
H = 39. In the upper p a i r of photographs, the dye is in j ec t ed through
one of t h e s l i t s (pos i t ion 4, Fig. 3.13, l e f t ) , whereas i n the lower
p a i r the dye i s in jec ted downstream f rm a s l i t (pos i t ion 3) . The f i e l d
i s seen t o have a s ign i f i can t s t a b i l i z i n g influence. I n f a c t , some sta-
b i l i z a t i o n , espec ia l ly i n the i n t e r i o r of the flow f i e l d , w a s observed
Figure 3.15 i l l u s t r a t e s the e f f e c t of the f i e l d when NRe e
f o r NR = 53, NRee = 5800 (H = 39, s = 0.27) . er It should be noted t h a t t he use here of two t angen t i a l s l i t s t o
drive the vortex i s e spec ia l ly conducive t o generation of i n s t a b i l i t i e s
i n the boundary layer because of the la rge asymmetric ve loc i ty per tur -
bat ions associated with t h e j e t flow. The f a c t t h a t , even i n t h i s un-
favorable case, t he magnetic i n t e rac t ion i s able t o s t a b i l i z e the flow
t o a s ign i f i can t extent is encouraging. Vortexes generated by multi-
pored in jec t ion as' previously discussed (page 9) should be more e a s i l y
s t ab i l i zed .
Figure 3.16 includes photographs depict ing flow i n the end w a l l
boundary layer f o r a main flow t angen t i a l Reynolds modulus (i .e . , outside
the end w a l l boundary l a y e r ) of 425, which i s s l i g h t l y above the maximum
observed t r a n s i t i o n point without an applied f i e l d . Here t h e dye i s in-
jec ted in to the boundary l aye r near t he periphery (as indicated i n Fig.
3.14, r i g h t ) . Note the i n s t a b i l i t y i n the dye t r a c e f o r
(Fig. 3.16, l e f t ) . When t h e a x i a l magnetic f i e l d ' is applied, giving
H = 31, S = 2.2 (Fig. 3.16, r i g h t ) , s t a b i l i z a t i o n of t h e boundary-layer
flow i s evident. Note p a r t i c u l a r l y the suppression of the dye t r ace
f luc tua t ions and the increase i n the t angen t i a l ve loc i ty r e l a t ive t o the
r a d i a l ve loc i ty . Analysis of these photographs ind ica tes , f o r example,
t h a t t he r a t i o of t angen t i a l t o r a d i a l ve loc i ty i s increased from approxi-
mately 1 t o approximately 12 by appl ica t ion of the f i e l d .
ve loc i ty r a t i o f o r t h e main flow (based on t o t a l volumetric flow r a t e
and j e t area and assuming two dimensionality) i s about 35 . Stab i l i za t ion
of t he end wall boundary layer should e f f e c t i v e l y reduce the sho r t - c i r -
c u i t i n g r a d i a l mass flow as discussed i n Refs. 37 and 40.
= - S = 0
The average
71
"I cn W ZX
0
72
I
73
The e f f e c t of t he magnetic f i e l d on t h e s t rength of t h e vortex w a s
invest igated semiquant i ta t ively by the dye pulse in j ec t ion technique. I
A quantity ind ica t ive of the effect iveness of vortex generation is a, t he j e t ve loc i ty recovery r a t i o previously considered. For t h i s case
a Qo/Qj, where Q i s the t angen t i a l per iphera l ve loc i ty (measured a t
a reference radius r a t i o r'= r/r
of t he i n l e t j e t . Figure 3.17 i s a graph of t he observed a values as
a function of r a d i a l Reynolds modulus, NRer (= Qr r p/p = m/2 1~ p), f o r
t he case of no magnetic f i e l d , Bo = 0 (H = 0) , and f o r the case of an
applied axial magnetic f i e l d , Bo = 62 kilogauss (H = 39 and 41) . t h a t H (= Bo ro ,/- ) is the Hartmann modulus defined i n terms of the
tube radius, r .§§
The spread between these runs i s not bel ieved t o be a consequence of
the small difference i n temperature, but r a the r an indicat ion of t he
la rge experimental uncer ta in t ies .
from the theory of R ~ s e n w e i g ~ ~ assuming laminar flow f o r two assumed
values of the parameter, h (= r./r , where r
of j e t en t ry ) .
w a s observed t o occur at NRe
Note t h a t t h e measured curve f o r 200"F, (H = 0) i s near ly p a r a l l e l t o the t h e o r e t i c a l laminar curve for h = 0.9 up t o NRe
t r a n s i t i o n t o a much f l a t t e r curve of a versus NRe r
i n the range 8 < N R ~ This t r a n s i t i o n near ly concides with v i sua l r
observations of t he onset of i n s t a b i l i t y .
i s applied, t he measured curves of a versus NRe
t h e t h e o r e t i c a l laminar curve up t o NRe
(210°F).
a l l y observed a t 210°F i s 1580, N R e ' = 17, which i s j u s t below the t r an - r
s i t i o n region i n t h e a curve. A magnetic f i e l d s t rength s u f f i c i e n t t o
give H = 41 has therefore increased t h e t r a n s i t i o n Reynolds modulus a t
210°F by a f ac to r of about 4.3.
0
= 0.8), and Q is t h e average ve loc i ty Q 0 j
Note
Data f o r two runs (200°F and 210°F) a re included. 0
The dashed curves a re ca lcu la ted
i s t h e e f f ec t ive radius J O
With H = 0,- onset of t h e i n s t a b i l i t y described on page 66
7.0 Y = 370 a t 210°F, corresponding t o NRe e r
'v = 8, and t h a t
appears t o occur r
< 13. When t h e 62-kilogauss f i e l d
a re near ly p a r a l l e l t o r
2 13 (200°F) and t o N R ~ r r 1 1 7
The t angen t i a l Reynolds modulus a t which t r a n s i t i o n w a s visu- Y
Figure 3.18 summarizes the observed increase i n j e t ve loc i ty recovery
as a funct ion of t he magnetic i n t e rac t ion parameter, r a t i o a t constant NRe r
'The curves i n Figs . 3.17 and 3.18 were f a i r e d through t h e data
" J f i s proport ional t o t h e r a t i o of magnetic t o viscous forces .
by eye.
UNCLASSIFIED
. 0.7
0.6
0.5
0.4
0 4 8 12 16 20 24 28 32 36 40
Fig. 3.17. Effect of a 62-Kilogauss Magnetic F ie ld on J e t Velocity Recovery Ratio for Vortex-Type Flow.
. .
1.5
1.4
1 . 3
1.2
1.1
1.0
. 0 ' J
0 1 2 3 4 5
Magnetic Interaction Parameter, S
Fig. 3.18. Increase i n J e t Velocity Recovery Ratio a t Constant Radial Reynolds Modulus Correlated with Magnetic In te rac t ion Parameter.
2 6 S ( = u B For la rge S, there i s l i t t l e e f f e c t of t he f i e l d
on a, since Qo, and hence N R ~ , is small; and the flow i s s t ab le without
an appl ied f i e l d . A s Qo is increased and S correspondingly decreased,
t he flow becomes unstable; and t h e magnetic f i e l d i s e f f ec t ive i n pro-
ducing s t a b i l i t y f o r S grea te r than about 1. Note t h a t when S i s l e s s
than 1 the flow i s unstable, but a is s t i l l increased by t h e magnetic
f i e l d .
of about 1 .4 . estimates of sk in - f r i c t ion coe f f i c i en t s were made from the observed a values .
s t r e s s i s decreased by a f a c t o r of 2 t o 2.5 by the appl ica t ion of a
magnetic f i e l d t o give H 40, S 1.
ro /p a,).” 0
e
For 1 < S < 1.25 ( s t a b i l i z e d flow), a i s increased by a f ac to r
Using the torque balance ana lys i s described earlier;2
Y This ana lys i s ind ica tes t h a t , f o r NRe = 1800, the w a l l shear e
Figure 3.19 graphs the observed va r i a t ion i n the t angen t i a l Reynolds * modulus a t t r a n s i t i o n t o i n s t a b i l i t y , N R ~ ~ , with the square of t he H a r t -
mann modulus, H2.“
mentally by varying Bo while maintaining a/p constant . The data suggest
Note t h a t t h e va r i a t ion i n H2 w a s obtained experi-
a more or
As a
parame t e r
formed a t
giving H~
* l e s s l i n e a r increase of NRe
fu r the r t e s t of t he v a l i d i t y of H2 as a c h a r a c t e r i s t i c magnetic
influencing the t r a n s i t i o n phenomenon, experiments were per-
77°F where u/p i s about 204 of the corresponding value a t 210°F,
= 350 f o r Bo = 62 ki logauss .
with H2 f o r H2 > 200. e
The t r a n s i t i o n Reynolds modulus * w a s observed t o be 730 under these condi t ions.
found t o be 425, i n agreement with t h e corresponding value a t 210°F as p l o t t e d i n Fig. 3.19. t o give H2 = 350, a value of NRe
as compared with the value 730 observed a t 77°F f o r H2 = 350.
For H2 = 0, NRee was
It is s ign i f i can t t h a t by decreasing Bo at 210°F *
of about 680 i s read from Fig. 3.19, e
It i s c onc.luded from these i n i t i a l experiments t h a t t he appl ica t ion
of an axial magnetic f i e l d t o je t -dr iven vortex flow of an e l e c t r i c a l l y
conducting l i q u i d is e f f ec t ive i n suppression of t he boundary-layer i n s t a -
b i l i t y on the concave wall which appears t o i n i t i a t e general turbulence
throughout t he flow f i e l d . A s i g n i f i c a n t increase i n t h e recovery of
‘ S i s proport ional t o the r a t i o of magnetic t o i n e r t i a l forces . 66 “The curve through a l l t h e da ta poin ts f o r t he two runs w a s
obtained by a third-degree polynomial l e a s t -squares ana lys i s .
I I .
UNCLASSIFIED
Fig. 3.19. Variation i n Transit ion Reynolds Modulus with Hartmann Modulus Squared, Obtained by Variation i n Magnetic Field Strength. 210°F data .
78 . i n j e c t i o n ve loc i ty as tangent ia l ve loc i ty i s thereby ef fec ted .
axial magnetic f i e l d i s a l s o e f fec t ive i n suppression of boundary-layer
i n s t a b i l i t y on t h e end w a l l , thus increasing the r a t i o of tangent ia l t o
radial ve loc i ty . These r e s u l t s are s u f f i c i e n t l y encouraging t o j u s t i f y
continued e f f o r t i n t h i s a rea . For example, it is proposed t o extend the range of Hartmann modulus by higher temperature operation, by the
use of higher magnetic f i e l d strength, ' and by increasing t h e s i z e of
the vortex chamber.
The
3.2. Turbulent Transport Studies
R . P. Wichner
The equations governing the mean ve loc i ty f i e l d i n turbulent , incom-
press ib le flow are the Reynolds equations which may be wri t ten:
aVi avi 1 ap a av av axi a t axi .axi ax
- + + * - - - - - - + - [ v (2+A)-s]. (3.8)
i 3
The c a p i t a l l e t t e r s represent averaged values.
values of the products of the f luc tua t ing i and j components of t h e
ve loc i ty) are the Reynolds s t r e s s e s .
bution t o t h e momentum conductivity of t h e f l u i d over and above the
molecular conductivity,
The terms vivj ( the mean
These represent the eddy c o n t r i -
v(%+5) . ax axi
j
Since the molecular conductivity i s known i n terms of ve loc i ty gradients -
t h i s i s simply .the Navier-Stokes s t r e s s tensor - the above equation may
be solved f o r the mean ve loc i ty when the eddy conductivity i s absent; i . e . ,
i n laminar flow.
i s knoim, hence it i s impossible t o solve f o r the turbulent ve loc i ty
f i e l d .
No equivalent expression f o r the turbulent conductivity
'An 80-kilogauss solenoid i s avai lable i n the ORNL Magnet Laboratory.
79
Although the re have been a few notable attempts a t theo re t i ca l
analysis of t he Reynolds s t r e s s problem, none has succeeded. It appears
t h a t i f success is t o be achieved i n t h i s approach t o the turbulence
problem, it w i l l r e s u l t from the examination and ana lys i s of a body of
Reynolds s t r e s s da ta from a va r i e ty of flow geometries.
U n t i l t he development of t he hot-wire anemometer i n the 1 9 3 0 ' ~ ~ it w a s not possible t o measure these eddy s t r e s s e s . Since then, some l imi ted
measurements on a few important geometries have been made; however, there
has been no systematic study i n t h i s f i e l d , and the avai lable experimen-
t a l data on Reynolds s t r e s s e s i s inadequate t o form a guide f o r a general
theory. Therefore, a Reynolds s t r e s s measurement e f f o r t is being made which has thus f a r progressed through the following s t eps :
1. An improved theory r e l a t i n g the s igna l of a hot-wire anemometer
The' new method requires t o the Reynolds s t r e s s e s has been formulated.42
a l i n e a r l i z e d response anemometer and divulges the complete s t a t e of
turbulent stress a t a point from a s ingle experiment. Older methods
required t h a t a separate a t t ack be made on each of t he six components
of t h e s t r e s s . In addi t ion, the proposed method has the capab i l i t y of
eliminating the need f o r the usual low-turbulence i n t e n s i t y approximation,
SO t h a t flows which possess mean-square turbulent ve loc i ty f luc tua t ions
a t l e a s t as great as 1 6 of the square of the mean ve loc i ty can be studied.
2.
t o more d i r e c t l y r e l a t e t he experiment t o pro jec ts of i n t e r e s t a t ORNL. A water loop has been fabr ica ted i n which t e s t sec t ions having various
flow geometries may be placed (Fig. 3.20). The s t r ingen t f l u i d c l e a n l i -
ness, deaeration, and temperature s t a b i l i t y requirements of water hot-
wire anemometry have, after some d i f f i c u l t y , been met. I n addi t ion,
extraneous flow f luc tua t ions have been eliminated.
It w a s decided t o perform the measurements using water i n order
3. d i f f i c u l t y .
Select ion of an appropriate hot-wire anemometer presented. some
Most commercial anemometers a re designed f o r use i n a i r and
42R. P. Wichner and F. N. Peebles, "Determination of t he Six Reynolds Stresses by t h e Linearized-Response Hot -Wire Anemometer, " p . 361 i n Developments i n Theoret ical and Applied Mechanics, Vol. 1, Plenum Press, New York, 1963.
80 . .
81
'8
do not have adequate wire heating r a t e s f o r water use. In addi t ion, it
w a s required that t h e anemometer output s igna l be l i nea r i zed with respect
t o f l u i d ve loc i ty . It w a s , therefore , necessary t o design and f ab r i ca t e
an instrument t o meet t h e requirements of t h i s experiment; t h i s system i s
shown i n the photograph and schematic of Fig. 3.20: The anemometer i s
b u i l t around an E'lectronics Associates TR-5 analog computer (Fig. 3 .21) .
The l i nea r i za t ion i s accomplished by summing and squaring components of
t he computer. It can be shown t h a t f o r t h e condition of the experiment,
e l e c t r i c a l compensation f o r the thermal i n e r t i a of t he wire i s not neces-
sa ry .
4. of which were designed f o r use i n a i r or with a p a r t i c u l a r instrument,
it w a s decided t o f ab r i ca t e these key items. Some of t h e handiwork i s
shown i n Fig. 3.22; t he support pins a re sewing needles, and the wire i s
0.0002-in. -diameter nickel . The probes y i e l d an adequately s t ab le read-
ing and can withstand a ve loc i ty of 1 5 f t / s e c .
After t e s t i n g some commercially ava i lab le hot-wire probes, most
5. Some t y p i c a l ca l ib ra t ion curves - p l o t s of anemometer output
voltage versus ve loc i ty - are shown i n Fig. 3.23. These indicate t h a t
l i nea r i za t ion of the anemometer response has been s a t i s f a c t o r y . The high
s e n s i t i v i t y t o ve loc i ty i s p a r t i a l l y due t o the high heating r a t e (a cur-
ren t of 0.175 amperes) which c rea t e s i n the wire a power densi ty of lo7 kw/ l i te r .
6 . Some preliminary r e s u l t s for a t e s t a t a Reynolds modulus of
158,000 are shown i n Fig. 3.24. i n water f lows i s possibly l e s s than an t i c ipa t ed based on measurements
i n a i r .
It i s seen that t h e l e v e l of turbulence
43
4 3 J O h n Laufer, "The Structure of Turbulence i n Ful ly Developed Pipe Flow, NACA-1174, 1954.
Constant Volt age
0 7 a b
Fluctuat ing Lsnplif ,?r
Power Supply n \ J 0 0 I,
Signal Linear - i za t ion
E
I Vo It a ge Squared
UNCLASSIFIED DWG. 64-7781
Analog Computer (TR-5)
- Averaging Device
Fig. 3.21. Linearized Response, Constant Current Hot-wire Anemometer System Schematic.
L--- ------ b
03 Iu
- J
Voltage Linear with Velocity
Mean Square Fluctuating
Pot e n t iome t e r Re c o r der
80
j k
a, -P
cd 17 .d
m a, v1 v1 a, k
-P
u1
.
83
84
a -
7-
6 -
5 -
4 -
3-
2 -
1-
0-
UNCLASSIFIED ORNL DWG. 64-7782
Fract ional Distance from W a l l t o Centerline
1.0 0.82 0.63
0.25 0.096
0.44
8- 8-
7- 8
7- 6- 7
6 - 5- 6
5 - 4- 5
4- 3' 4
3- 2-
2-
1- 2
1-
Flow - 1-30 gal/min
NRe - 158,000
0-
Ordinate Corresponds t o Posi t ion Number -
A B C D E F
1 0-
c
Posi t ion No.
A B C D E F
I
Posi t ion i n
0 2 4 6 8 10
Anemometer Output (Volts)
Fig. 3.23. Linearized Response Calibration Curves, Run LT6.
c
85
0.12
0.1
0.08
0.06
0.04
0.02
0 0.2 0.4 0.6 0.8
Fract ional Distance from Wall t o Centerline
1 .o
Fig. 3.24. Turbulence Pressure vs Posi t ion i n Pipe.
86
3.3. Boundary-Layer Transient Phenomena
G. J. Kidd, Jr.
Previous studies44, 47 a t ORNL have indicated t h a t cyc l ic s t r e s s e s
induced i n a s o l i d by thermal f luc tua t ions i n a f l u i d flowing pas t the
s o l i d surface can cause fatique-type f a i l u r e s and/or accelerated cor-
rosion i n t h e s o l i d . The extent of t h i s damage w a s found t o be a func-
t i o n of such f a c t o r s as the amplitude of the temperature f luc tua t ions ,
t h e i r frequency spectrum, the length of exposure, and the r e l a t i v e physi-
c a l and thermal propert ies of the f l u i d and s o l i d . Since moderatly high-
frequency f luc tua t ions (0.1 t o 10 cycles/sec) a r e normally generated
continuously during the operating l i fe t ime of a reactor , a s u f f i c i e n t
number of cycles can be accumulated t o produce fatigue-type f a i l u r e s
even a t low amplitudes.
These thermal t r a n s i e n t s a r e a r e s u l t of t h e random motion of the
f l u i d p a r t i c l e s ; thus, even i n a geometry as simple as a tubular f u e l
element, i n t e r n a l l y cooled, turbulent eddies penetrate the viscous sub-
layer adjacent t o the bounding surface and cause rap id f luc tua t ions i n
the l o c a l r a t e of heat removal. I n entrance regions, diverging channels
o r downstream of obstructions, t r a n s i e n t s of la rge amplitude can be ex-
pected.
by the p o s s i b i l i t y of adverse temperature p r o f i l e s i n the f l u i d streams.
I n addition, the combination of a good thermally conducting f l u i d i n
contact with a low-conductivity w a l l i s espec ia l ly l i k e l y t o r e s u l t i n
For c i rcu la t ing- fue l reactors , the problem is f u r t h e r i n t e n s i f i e d
44J. J. Keyes, Jr. and A. I . Krakoviak, "High-Frequency Surface Thermal Fatigue Cycling of Inconel a t 1405°F," Nuclear Sei . Eng., 9(4): (April 1961).
4sR. A. Suehrstedt e t a l . , "Applications of Concepts of Pentration Theory t o Heat Transfer," USAM: Report KT-378, MIT Pract ice School, December 1958.
a Surface and a Flowing Fluid," USAEC Report KT-397, MIT Pract ice School, May 195 9
47H. W. Hoffman e t a l . , "Fundamental Studies i n Heat Transfer and Fluid Mechanics - Status Report Ju ly 1, 1959 - February 29, 1960,~~ USAM: Report O R N L - C F - ~ O - ~ O - ~ , Oak Ridge National Laboratory, October 4,
46A. E . Higinbotham e t a l . , "Study of Unsteady Heat Transfer Between '
1960.
87
l a rge surface temperature o s c i l l a t i o n s . These thermal osc i l l a t ions w i l l ,
i n many instances, p e r s i s t beyond the confines of the reac tor core and
can cause damage i n connecting l i n e s and ex terna l heat exchangers.
In t h a t t h e mechanism of thermal t r a n s i e n t generation a r i s e s from
the e f f e c t s of turbulence near a wall, an understanding of the nature of
turbulent i n s t a b i l i t i e s i n boundary layers i s needed. A t present , t he
theo re t i ca l and experimental knowledge of such cha rac t e r i s t i c s of t he
flow as eddy s i z e and ve loc i ty d i s t r i b u t i o n near a wall is meager.
Several recent inves t iga t ions 47-51 concerned pr imari ly with mass t r a n s f e r
have suggested t h a t t he viscous sublayer i s , i n f a c t , in te rmi t tan t , break-
ing up per iodica l ly i n t o turbulent eddies o r a t l e a s t being thinned suf-
f i c i e n t l y t o permit eddies from the bulk of the f l o w t o approach the w a l l . This inves t iga t ion w i l l attempt t o answer, through the quant i ta t ive study
of turbulent ve loc i ty and temperature p r o f i l e s a t a s o l i d surface, some
of t he questions concerning therma.1 i n s t a b i l i t i e s r e l a t e d t o conditions
and geometries of i n t e r e s t i n reac tor design.
The experimental system t o be used i n t h i s study i s shown p i c t o r i a l l y
i n Fig. 3.25 and diagrammatically i n Fig. 3.26. A constant head, once
through, deaerated water supply system i s u t i l i z e d t o provide a s teady
f low, f r ee from f luc tua t ions from pumps, valves, e t c . The constant head
tanks a re approximately 2 f t i n diameter and 1 5 f t t a l l ; flow i s adjusted
by changing the e leva t ion of e i t h e r o r both of t h e overflow l i n e s . A
200-gpm cold water (vacuum) deaerator has been i n s t a l l e d and i s i n opera-
t i o n .
gas concentration of l e s s than 0.05 c c / l i t e r .
deaerated water a r i s e s from the f a c t t h a t the s igna l c rea ted by a gas
bubble flowing pas t one of the sensing elements is indis t inguishable from
t h a t produced by an eddy of cool f l u i d .
The deaerator suppl ies water t o the t e s t sec t ion w i t h a dissolved
The necess i ty f o r using
48P. V. Dankwerts, Ind. Eng. Chem., 43: 1460 (1951). -P. Harr io t t , "A Random Eddy Modification of t h e Penetration Theory,
'OL. P. Reiss and 'T. J. Hauratty, "Measurements of Instantaneous
Chem. Eng. Se i . , 17: 149-154 (1962).
Rates of Mass Transfer t o a Smal l S i n k on a Wall," A.1.Ch.E. Journal, 8 ( 2 ) : 245-247 (May 1962).
"J. Sternberg, "A Theory f o r the Viscous Sublayer of a Turbulent Flow," Fluid Mech., 13: 241-271 (1961).
88
I
i
. $9
C n
The t e s t sec t ion i l l u s t r a t e d i n Fig. 3.27 has been used t o inves t i -
gate severa l design concepts and t o check out t he instrumentation. A
number of qua l i t a t ive experiments have a l s o been performed. There a re
20 "Calrod" 5-kw heaters embedded i n an aluminum cas t ing , surrounding
a sec t ion of 2-in. schedule 40, type 347 s t a i n l e s s s t e e l p ipe .
i s 10-ft long with t h e heater encompassing the last 6 f t .
type thermocouples a re located 1 f t from the o u t l e t .
gained from t h i s t e s t sec t ion has demonstrated t h a t t h i s type of thermo-
couple can be fabr ica ted and i n s t a l l e d t o very c lose to le rances so t h a t
meaningful data can be obtained. For example, ana lys i s of t h e thermal
emfs by means of an analogue f i l t e r i n g device using the techniques of
frequency domain ana lys i s w a s used t o compute t h e preliminary power spec-
t r a l densi ty (PSD) of t he surface temperature f luc tua t ions which i s p l o t -
t e d i n Fig. 3.30. Here the PSD is a s t a t i s t i c a l quant i ty defined as the
mean-square f luc tua t ion per u n i t bandwidth. Note t h a t most of t h e t r an -
s i e n t thermal power per u n i t bandwidth i s i n the range of frequencies
below one cps; these low-frequency thermal o s c i l l a t i o n s a re bel ieved t o
be s ign i f i can t from the standpoint of thermal f a t igue .
The pipe
Four gunbarrel-
Operating experience
A second, more sophis t ica ted , . t e s t sec t ion now under construct ion
is shown i n Fig. 3.28. l e s s s t e e l , 8-ft long with an ins ide diameter of 2 i n . and an outside
diameter of 5 i n . and w i l l contain 16 gunbarrel thermocouples placed along
It i s made from a s o l i d piece of type 321 s t a i n -
the e n t i r e length of t he channel. There are th ree access po r t s i n the
sec t ion through which hot-fi lm anemometer probes and/or stream tempera-
ture probes can be inser ted . The heaters a re spec ia l ly shaped Chromalox
rods designed t o provide grea te r heating a rea per un i t length; they a re
brazed i n t o matching grooves along t h e f i n a l 6 f t of the t e s t sec t ion .
The e l e c t r i c a l power supply f o r t h e heaters i s a 472/550 v, 400/467 kw
Westinghouse dc generator . Direct cur ren t i s necessary t o keep t h e back-
ground ac noise l e v e l as low as possible s ince the s igna l s produced by
t h e thermocouples a re very small (of t h e order of 10 t o 100 microvol ts) .
A simple mathematical model of t h e temperature d i s t r i b u t i o n and
heat t r a n s f e r across the boundary l aye r has been developed5' which takes
52J. J. Keyes, Jr., "Penetration Theory of Transient Heat Transfer Between a Surface and a Fluid i n Turbulent Flow," t o be published.
-. - 91
.
.
UNCLASSIFIED ORi'lL-LR-DWG. 70259
Fig. 3.27. Test Section for Study of Temperature Fluctuations at a Heat-Transfer Surface.
I -.I a
92
. .
. 93
i n t o account the breaking up
This model has been analyzed
t h a t the preliminary r e s u l t s
of the viscous sublayer and eddy penetration.
using an analog computer, and it w a s found
obtained from experiments on the f i r s t t e s t
sect ion were q u a l i t a t i v e l y consis tent with the simple theory. It i s a n t i -
c ipated t h a t more accurate data w i l l be obtained with t h e new t e s t sec t ion
and improved instrumentation; t h i s w i l l provide a b e t t e r bas i s f o r ref ine-
ments i n the a n a l y t i c a l model.
Assembly and c a l i b r a t i o n of an integrated instrumentation system
(Fig. 3.29) f o r recording and analyzing the complex t rans ien t s igna ls
generated by the temperature and ve loc i ty (hot-film anemometer) probes
i s e s s e n t i a l l y completed. This system w i l l enable d i r e c t multichannel
recording of s igna ls as low as 10 microvolts peak-peak i n t h e frequency
range from 0 t o 2000 cps. The read-out system cons is t s of a low-fre-
quency spectrum analyzer with provision f o r d i r e c t amplitude-frequency,
t r u e r m s amplitude-frequency, or power s p e c t r a l density-frequency scan-
ning, a t y p i c a l spectral-densi ty p r o f i l e is shown i n Fig. 3.30. This
instrument w i l l analyze s igna ls i n the frequency range from about 0.5
t o 2000 cps d i r e c t l y with resolut ion of as low as 0.3 cps. Using an FM carr ier- type tape record-reproduce system f o r frequency mult ipl icat ion,
t h e system w i l l analyze s igna ls with frequencies as low as 0.05 cps.
A hot-film surface probe has been developed which has enabled d i r e c t
determination of t h e instantaneous r a t e of heat t r a n s f e r and of the in-
stantaneous coef f ic ien t of heat t r a n s f e r a t t h e surface of the t e s t sec-
t i o n . It cons is t s of a v e r y t h i n s t r i p of gold O.OO5-in. wide by 0.250-
in, long vacuum deposited across a pair of platinum electrodes c a s t i n
epoxy res in . The surface of t h e probe i s accurately contoured t o f i t
f lush with the inside w a l l of the t e s t sec t ion . By u t i l i z i n g one of the
c i r c u i t s of the constant temperature hot-film anemometer t o power the
probe and by analyzing the voltage required t o maintain t h e constant f i l m
temperature by means of t h e power s p e c t r a l densi ty analyzer, the spectrum
of instantaneous surface heat f lux can be recorded. This i s a l so t h e
spectrum of t h e instantaneous heat- t ransfer coeffic5ent defined by the
re la t ion ,
I .
UNCLAS S I?? IED ORNL-LR-DWG. 74425
I I Oscillographic Recorder
Input from Surface Thermocouple Preamplifier D.C. Amplifier
r
r v Analyzer Recorder 0.5 - 1500 cps
I -____ -1
Record bw-Gain D.C. I I
Reproduce Tape Recorder
Analogue Computer
Power Spectral Density Analyzer
Fig. 3.29. Instrumentation Diagram for Analysis of Transient Thermal EMFS.
. I
95
- 250 200
i
3 n 50 a m
L
‘0 too -
z
z 2 10 W 3 2 5 W
I-
-J W
z . a
a i C
UNCLASSIFIED ORNL-LR-DWG. 74090
3EYNOLDS MODULUS = 73000
! 0.5 1 5 10 20 FREQUENCY, cycleslsec
Fig . 3.30. Preliminary Power Spectral-Density Analysis of Surface Thermal Instabilities for Turbulent Flow of Water.
.
.
since A t i s maintained constant by the e iec t ronic c i r c u i t r y .
intended t o compare the h(0) spectrum with t h e t ( 0 ) spectrum [where t ( 0 )
i s the instantaneous surface temperature a t an adjacent point i n a heated
t e s t - s e c t i o n wall as obtained’by a surface thermocouple] i n an e f f o r t t o
del ineate more c l e a r l y the nature of the boundary-layer f luc tua t ions
giving r i s e t o these t r a n s i e n t surface phenomena.
It i s
.
.
.
97
4. THERMOPHYSICAL PROPERTlES
4.1. Alka l i Liquid Metals
J. W . Cooke
53J. W . Cooke, "The Experimental Determination of the Thermal Con- duc t iv i ty of Molten Lithium from 600 t o 1550 Degrees Fahrenheit," USAM: Report ORNL-3390, Oak Ridge National Laboratory, January 1964.
of Molten Lithium from 320" t o 830"c,~~ J. Chem. Phys., bO(7): 1902-1909 54J. W . Cooke, "Experimental Determination of t he Thermal Conductivity
(April 1, 1964).
Metals and Alloys," p . 10-11 i n Problems i n Heat Transfer, ed. by M . A: Mikheev, Publishing House of the Academy of Sciences USSR, Moscow, 1959.
55N. A . Nikol ' sk i i e t a l . , "Thermal and Physical Propert ies of Molten
4 .1 .1 . Thermal Conductivity of Lithium
The measurement of t he thermal conduct ivi ty of molten l i thium
(99.82 wt % L i ) a t temperatures t o 1550°F has been completed; d e t a i l s
of the apparatus and of the measuring procedure a re t o be found i n Refs.
53 and 54. Final r e s u l t s (Fig. 4 .1 ) can be expressed by the equation
k = 19.76 (1 + 5.01 x t ) , (4 .1)
between 600" and 1500°F, where k i s i n Btu/hr.ft."F f o r t i n OF.
de t a i l ed e r r o r ana lys i s indicated the accuracy of t he da ta t o vary from
+7% a t the lower temperature t o +l$ a t t h e upper end of the range.
A
The present r e s u l t s a re compared i n Fig. 4.2 with predicted and
The unpublished measure- experimental values by other inves t iga tors .
ments of Mil ler and Ewing (Naval Research Laboratory) i n the temperature
range between 540" and 770°F f a l l roughly p a r a l l e l t o , but 4% below
the current da ta . Nikol ' sk i i and ~ o - w o r k e r s ~ ~ (1959) found a l e s s e r
'
dependence of t he thermal conductivity on the temperature. A t t he
higher temperatures, a mean l i n e through t h e i r data l i e s 1% below the
ORNL r e su l t s ; however, t he t o t a l s c a t t e r i n the Nikol ' sk i i da ta i s such
e
0
ln
0
ln
0
d- m
m
cu cu
( j o * 4 4 * J
4 /n + 8 )
A 1 I A I 1
3 n a N 0 3
1 V W tl3 H
1
a
99
I I
WIEDEMANN, FRANZ, LORENZ, AND SOMMERFELD EO.
UNCLASSIFIED ORNL- LR-D WG 7084 7
400 600 800 4000 rzoo (400 4600 TEMPERATURE ( O F )
Fig. 4.2. of Molten Lithium.
Comparison of Measurements on the Thermal Conductivity
100
as t o embrace the present data. Most recent ly , Rudnev e t a ~ ' ~ (1961) completed measurements on the thermal d i f f u s i v i t y of l i th ium from 653" t o 1845°F. of 4% above the current r e s u l t s .
the e a r l y data of Webber e t a ~ ' ~ (1955) are i n s u b s t a n t i a l disagreement
i n regard t o both magnitude (except at t h e low l i q u i d temperatures) and
dependency on temperature with a l l l a t e r r e s u l t s .
Thermal conduct ivi t ies derived from t h i s data f a l l a maximum
Final ly , it i s seen i n Fig. 4.2 t h a t
I n Fig. 4.2 the present data a r e a l s o compared with predict ions
based on the cor re la t ion of Ewing e t and t h e t h e o r e t i c a l expression
of Wiedemann and Franz" as modified by Lorenz" and l a t e r by Somerfeld"
(using recent e l e c t r i c a l r e s i s t i v i t y values determined by Kapelner62) . The agreement i s good i n respect t o the slopes of the curves, but the
predicted values f a l l lo$ and 6%, respectively, above the ORM; data .
Further, using t h e r e s u l t s of B i d ~ e l l ~ ~ f o r the thermal conductivity of
the s o l i d l i thium, agreement i s found with the t h e o r e t i c a l expressions
of M ~ t t ~ ~ and Rao6' t h a t the r a t i o of the thermal conduct ivi t ies of a
561. I. Rudnev, V. S. Lyshenko, and M. D. Abramovich, "Diffusivi ty of Sodium and Lithium," Atomnaya Energiya, 11: 230-232 (September 1961); see Atomic Energy, 3 : 877-880 (March 1962), t r a n s l a t e d by Consultants Bureau.
of Molten Lithium," Trans. Am. SOC. Mech. Engrs., 77: 97 (1955). 57H. A. Webber e t a l . , "Determination of the Thermal Conductivity
58c. T . Ewing e t a l . , "Thermal Conductivit ies of Metals ,If C h e m Eng. Prog. Symposium Series , No. 20, 23: 19-24 (1957).
59G. Wiedemann and R . Franz, "The Thermal Conductivit ies of Metals," Ann. F'hysik. u. Chem., 89: 497-531 (1853).
'OL. Lorenz, "The Conductivity of Metals f o r Heat and E l e c t r i c i t y , " Ann. Physik, 13: 422-447 (1881).
"A. Sommerfeld, "The Electron Theory of Metals Based on Fermi S t a t i s t i c s , Z. Physik, 47: 1-32 (1928).
62S. N . Kapelner, "The E l e c t r i c a l R e s i s t i v i t y of Lithium and Sodium- Potassium Alloy," P r a t t and Whitney Aircraf t Report PWAC-349, June 30,1961.
63C. C . Bidwell, "Thermal Conductivity of L i and N a by a Modification of the Forbes Bar Method," Phys. Rev., 28, 58h-597 (1926).
64N. F. Mott , "The Resistance of Liquid Metals ,If Proc . Roy. Soc . (London), Series A, 146: 465-472 (1954).
.
.
"M. R . Rao, "Thermal Conductivity of Liquid Metals," Indian J. Phys. , 16: 155-159 (1942).
101
l i q u i d metal before and a f t e r melting should be equal t o the square of the r a t i o of the corresponding atomic frequencies. I n cont ras t t o r e -
s u l t s with sodium and potassium, t h e thermal conduct ivi ty of molten
l i th ium exhib i t s a pos i t i ve t r end with temperature (both experimentally
and t h e o r e t i c a l l y i n the temperature range of F ig . 4.2)
t o be r e l a t ed t o d i f fe rences i n t h e s t r u c t u r a l change which occurs a t
melting; and it has been noted t h a t at t he melting poin t the r a t i o of
so l id- to- l iqu id conduct iv i t ies is grea te r f o r l i th ium than f o r sodium
or potassium (ks/kL 3 1.8, 1.3, and 1 .6 f o r l i thium, sodium, and potassium,
respec t ive ly) .
This is bel ieved
4.1.2. Surface Tension of Potassium
Preliminary r e s u l t s f o r t h e surface tension, r e l a t i v e t o helium, of
99.96 w t % potassium have been obtained i n the temperature range 70 t o
3 0 8 " ~ using the maximum-bubble pressure apparatus i l l u s t r a t e d schemati-
c a l l y i n Fig. 4.3. High-purity helium w a s supplied a t very low flow t o
a c a p i l l a r y tube whose end w a s immersed t o various depths i n the l i q u i d .
The maximum pressure a t t a ined within the slowly growing bubble before
disengagement from t h e c a p i l l a r y t i p w a s measured with a micromanometer
( sens i t i v i ty , kO.01 mm) using Octo1 S f l u i d . The immersion depth w a s
es tab l i shed by a d i a l ind ica tor ( s e n s i t i v i t y , 20.025 mm) which w a s zeroed
a t the point of e l e c t r i c a l contact of the c a p i l l a r y with the l i q u i d sur- face. The surface tens ion can then be calculated, following the i t e r a -
t i v e technique of Sugden66 which accounts f o r the nonspherical shape of
the bubble, using t h e inside radius of t he c a p i l l a r y t i p , the maximum
bubble pressure, and the immersion depth. For these experiments, it has
been es tab l i shed t h a t t he simpler ca l cu la t ion using Schroedinger's
e quat ion, 67
"S. Sugden, "The Determination of Surface Tension from Maximum Pressure i n Bubbles," J. Chem. Soc.., 1: 858 (1922).
67G. Schroediner, "Note on the Capi l la ry Pressure i n Gas Bubbles," Ann. Physik, 46: 413 (1915).
102
DEPTH ADJUSTMENT
VACUUM - DRY BOX \
FURNACE
THERMOCOUPLES
UNCLASSIFIED ORNL-DWG 63-332f
/NEEDLE VALVE
I I 1 I I I
CAPILLARY TUBE
I
-0CTOL S
I. MICROMANOMETER . Fig. 4.3. Maxim-Bubble-Pressure Apparatus for Surface-Tension
Determinations with Liquid Metals.
gives values agreeing t o within a f r ac t ion of a per cent with those
obtained by the Sugden procedure. In Eq. 4.2, IJ i s the surface tension
(dyne/cm), r, the t i p radius (cm), and h, t he hydrostat ic heat of f l u i d
above the c a p i l l a r y t i p (em).
For t h e current measurements, the apparatus d i f f e red from t h a t des-
c r ibed above i n t h a t water, r a the r than Octo1 S, w a s used i n the manometer.
The water and potassium were separated by liquid-nitrogen-cooled molecu-
la r - s ieve t r aps . Some l o s s i n s e n s i t i v i t y i n t h e pressure measurements
resu l ted f r o m t h e e f f e c t of the compressible gas volume i n these t r a p s .
Further, it w a s found t h a t t he c a p i l l a r y t i p could be M e r s e d t o only
1/4 of i t s design depth; t h i s introduced an addi t iona l loss i n precis ion.
Final ly , s ince the helium pur i f i ca t ion system had not yet. been completed,
some contamination of the potassium possibly occurred; t he helium pur i ty
w a s of the order of 99.94'0 r a the r than the 99.99+% desired.
The da ta of t h i s preliminary measurement, p l o t t e d i n Fig. 4.4, can
be represented, using a least-squares f i t t i n g technique, by the equation,
(I = 118.5 - 0.'0998 t + 7.65 x t2 , (4 .3)
for t i n "C over t h e temperature range 70 t o 3 0 8 " ~ .
w a s 0.54 dyne/cm.
potassium temperature w a s decreasing a re cons is ten t with those f o r t he
i n i t i a l temperature-increasing per iod. A l s o shown i n Fig. 4.4 a re the
r e s u l t s of Quarterman and Primak6' obtained by a c a p i l l a r y r i s e technique
These inves t iga tors s t a t e t h a t t h e i r da ta may be low by as much as 15% due t o e r ro r s introduced i n t o the measurement of the magnitude of t h e
l i q u i d r i s e by uncer ta in t ies i n the l i q u i d meniscus.
The i-ms deviat ion
As noted i n Fig. 4.4, the data obtained while t he
Refinement of the apparatus i s not expected t o change the r e s u l t s
described by Eq. 4.3. This i s indicated by t h e r e s u l t s of a determina-
t i o n of t he surface tens ion of t r i p l y d i s t i l l e d , doubly deionized water
using the same apparatus arrangement as for t he potassium measurements.
"L. A. Quarterman and W. L. Primak, "The Capi l la ry Rise, Contact Angle, and Surface Tension of Potassium," J. Am. Chem. Soc., 72: 3035 (1950) -
UNCLASSlFl ED ORNL-DWG 63-3322
105
1 A QUARTERMAN AND PRIMAK ( R E F . 2 2 ) I I
200 2 5 0 300 350 400 450 500 550 600 TEMPERATURE (OF)
85
80 150
Fig. 4 . 4 . Surface Tension of Molten Potassium as a Function of Temperature.
I 4
It w a s found t h a t u f o r water against helium at 28"c w a s 2% below the
value reported by Dorsey6' f o r water r e l a t i v e t o a i r .
pointed out t h a t the major p a r t of t h i s 2% discrepancy can be a t t r i b u t e d
t o t h e e f f e c t s of using d i f fe ren t gases f o r t h e measurements.
Gambill7O
These s tudies w i l l be continued at higher temperatures a f t e r in -
corporation of t he changes necessary t o br ing the apparatus t o design
spec i f ica t ion .
4.1.3. Contact -Angle k t e r m i n a t ions
The extent t o which a l i q u i d "wets" a surface can be character ized
by the angle of contact between a l i q u i d droplet and the surface on which
the droplet r e s t s ; the angle of contact , 0, i s defined i n Fig. 4 .5 . From
thermodynamic consideration^,^^^^" it can be shown that
u = u - u c o s 8 , s a sv av (4.4)
where the subscr ip ts s, a, v indicate so l id , l i qu id , and vapor, respec-
t i v e l y , and i n combination designate t h e phase in t e r f aces .
exceeds u 8 w i l l be grea te r than 90 deg; t h i s i s a nonwetting condi-
t i o n .
8 < 90 deg.
When usa
sv , Conversely, if the l i q u i d spreads on the surface, usv > usa and
Since the wetting c h a r a c t e r i s t i c s of a l i q u i d aga ins t a surface a re
of importance t o a complete understanding (and predic t ion) of bo i l ing
and condensing phenomena, a program t o measure the contact angle f o r t he
a l k a l i l i q u i d metals i n contact with various surfaces has been i n s t i t u t e d
"N. E . Dorsey, Propert ies of Ordinary Water Substance, Reinhold Publishing Corp., New York, 1940.
7%. R . Gambill, "Surface Tension of Pure Liquids," Chem. Eng., 56: 146 (April 1958).
71W. D. Ha>kins, The Physical Chemistry of Surface Films, Reinhold Publishing Corp., New York, 1952.
72A. Bond:, "Spreading of Liquid Metals on Sol id Surfaces; Surface Chemistry of High-Energy Substances," Chem. Revs., 42(2) : 417-458 (1953).
106
t
t
UNCLASSIFIED ORNL-DWG 63-3323
VAPOR
VAPOR ( L'QU'" (@a/.: LIQUID
SOLID OSg SOLID use us"
( u ) NONWETTING (6) WETTl NG
Fig. 4.5. Shape of Sessile Drops for Wetting and Nonwetting Conditions.
107
using the sessi le-drop method. This technique, and a ca lcu la t iona l pro-
cedure f o r ex t rac t ing the surface tension from t h e measurements, have, been
discussed i n d e t a i l by Gi l l i l and . Preliminary experiments t o e s t ab l i sh
f e a s i b i l i t y i n t h e current appl icat ion have been completed with potassium
on a polished f la t , horizontal , type 316 s t a i n l e s s s t e e l p l a t e . Periodic
photographs of t he droplet were obtained while t he surface temperature
w a s increased s t e a d i l y t o 6 0 0 " ~ ; t yp ica l r e s u l t s a re shown i n Fig. 4.6. It w a s observed t h a t t he angle of contact var ied from an i n i t i a l value of
approximately 120-deg a t 70°C t o 0-deg a t 500°C some 95 minutes l a t e r .
A second droplet placed on the surface a t 1 8 0 " ~ produced an i n i t i a l con-
t a c t angle of 50 deg.
from evaporation of the droplet ( loss of mater ia l , obscuring of t he view
by the generated vapor, e t c . ) were encountered.
Above 6oo0c, experimental d i f f i c u l t i e s resu l t ing
Following completion of a more ref ined op t i ca l system, these experi-
ments w i l l be continued s o as t o d i f f e ren t i a t e t he influence of t he va r i -
ables time, temperature, and surface condition.
4.2. Miscellaneous Materials
4.2.1. E lec t r i ca l Res i s t iv i ty of Brass
W . R . Gambill
The temperature dependence of the e l e c t r i c a l r e s i s t i v i t y of seamless
70-30 brass tubing w a s determined over t he temperature range of 70 t o
500°C.
were i n good agreement. The mean values of r e s i s t i v i t y a re given i n
Table 4 . 1 .
The r e s u l t s obtained with two t e s t specimens of d i f fe ren t lengths
Table 4 .1 . E lec t r i ca l Res i s t iv i ty of Seamless 70-30 Brass
t ("c) 100 200 300 400 500
P, (P oh-cm) 7.39 8.29 9.37 10.58 11.75
7%. G . Gi l l i l and , "Investigation of the Wet tab i l i ty of Various Pure Metals and Alloys on Beryllium,'' USAM: Report om-3438, Oak Ridge National Laboratory, May 1963.
108
UNCLASSIFIED PHOTO 62327
( 0 1 TIME = Omin ( b ) TIME = 26min ( c ) TIME= 47 min TEMPERATURE = 447 F TEMPERATURE = 329O F TEMPERATURE = 445O F
( d ) TIME = 72 rnin (e) TIME = 95 min TEMPERATURE = 772O F TEMPERATURE = 904' F
Fig. 4.6. Sessile-Drop Shapes with Molten Potassium on a Stainless Steel Surface as a Function of Temperature and Time.
5 . BIBLIOGRAPHY
The information presented i n t h i s cur ren t report der ives , i n general ,
from s tudies i n i t i a t e d p r i o r t o t h i s repor t per iod. It seems des i rab le ,
therefore , t o l i s t ORNL repor t s and open- l i te ra ture publ icat ions issued
s ince the inception of these programs s o t h a t the reader i s able t o
acquaint himself conveniently with the f u l l scope of these s tud ie s .
Boiling Potassium Heat Transfer
1. W . R . Gambill and H . W . Hoffman, "Boiling Liquid-Metal Heat Transfer," American Rocket Society Paper No. 1737-61, May 1961.
2. H. W . Hoffman and A. I. Krakoviak, Forced Convection Saturat ion Boil- ing of Potassium a t Near-Atmospheric Pressure, pp. 182-203, "Proceed- ings of 1962 High-Temperature Liquid-Metal Heat Transfer Technology Meeting, I ' USAEC Report BNL-756, Brookhaven National Laboratory.
3. H . W . Hoffman, Studies i n Two-Phase Flow and Boiling Heat Transfer a t the Oak Ridge National Laboratory, pp. 18-23, "Two-Phase Flow Problems," - EURATOM Report, EUR 352.e, 1963.
4. A. I. Krakoviak, "Notes on t h e Liquid-Metal Boiling Process," USAM: Report ORNL-TM-618, Oak Ridge National Laboratory, Ju ly 10, 1963.
5. A . I . Krakoviak, Superheat Requirements with Boiling Liquid Metals, "Proceedings of 1963 High-Tempe r a tu re L i quid-Metal Heat Transfer Technology Conference, 'I USAM: Report 0 ~ ~ ~ 3 6 0 3 , Oak Ridge National Laboratory ( i n pub l i ca t ion ) .
6. H . W . Hoffman, Recent Experimental Results i n ORNL Studies with Boil- ing Potassium, "Proceedings of 1963 High-Temperature Liquid-Metal Heat Transfer Technology Conference , National Laboratory ( i n pub l i ca t ion ) .
USAM: Report ORNL-3603, Oak Ridge
General Boiling Studies
1. W . R . Gambill and R . D. Bundy, "Burnout Heat Fluxes f o r Low-Pressure Water i n Natural Circulat ion, ' I USAEC Report ORNL-3026, Oak Ridge .National Laboratory, December 20, 1960.
2. W . R . Gambill and R . D. Bundy, "HFIR Heat-Transfer Studies of Turbu- l e n t Water Flow i n Thin Rectangular Channels," USAEC Report ORNL-3079, Oak Ridge National Laboratory, June' 5, 1961.
3. W . R . Gambill, "A Survey of Boiling Burnout," B r i t i s h Chem. Eng., 93-98 (February 1963).
110
. n
4 . W . R . Gambill, "Generalized Predict ion of Burnout Heat Flux f o r Flow- ing, Subcooled, Wetting Liquids," Chem. Eng. Progr. Symposium Series , 59(41): 71-87 (March 1963); a l s o see W. R . Gambill, "Burnout Heat Flux Predict ion f o r Flowing, Subcooled, Wetting Liquids, pp. 112-137, Proceedings of Research Reactor Fuel Element Conference, Gatlinburg, Tennessee, September 1962, USAEC Report TID-7643 (Book I), 196.3,
Swirl-Flow Heat Transfer
1.
2 .
3.
4.
5 -
6.
7.
8.
W . R . Gambill and N . D. Greene, "A Study of Burnout Heat Fluxes Associated with Forced-Convection, Subcooled, and Bulk Nucleate Boil- ing of Water i n Source-Vortex Flow, '' USAM: Report ORNL CF-57-10-118, Oak Ridge National Laboratory, October 29, 1957.
W. R . Gambill and N. D. Greene, "A Preliminary Study of Vortex-Boiling Burnout Heat Fluxes, If J e t Propulsion, 28 (3) : 192-194 (March 1958).
W. R . Gambill and N. D. Greene, "A Preliminary Study of Boiling Burn- out Heat Fluxes f o r Water i n Vortex Flow," USAM: Report ORNL CF-58-4- 56, Oak Ridge National Laboratory, Apr i l 12, 1958; a l s o see A.1.Ch.E. Preprint 29, Second National Heat Transfer Conference, August 18-21, 1.958.
N. D. Greene and W. R . Gambill, "A Preliminary Inves t iga t ion of A i r Film Heat-Transfer Coeff ic ien ts f o r Free- and Forced-Vortex Flow Within Tubes.," USAFX: Report ORNL CF-58-5-67, May 23, 1958.
W. R . Gambill and N. D. Greene, "Boiling Burnout with Water i n Vortex Flow," Chem. Eng. Progr. , 54: 6&76 (October 1958).
W. R . Gambill, R . D. Bundy, and R . W. Wansbrough, "Heat, Burnout, and Pressure Drop f o r Water i n Swirl Flow Through Tubes with In t e rna l Twisted Tapes, 'I USAFX: Report ORN5;-29ll, Oak Ridge National Laboratory, March 28, 1960; see a l so Chem. Eng. Progr. Symposium Ser ies , 57(32): 127-137 (1961).
W . R . Gambill and R . D. Bundy, "An Evaluation of t h e Present Status of Swirl-Flaw Heat Transfer," ASME Paper No. 62-HT-42, Houston, Texas, August 1962; see a l s o USAM: Report ORNL CF-61-4-61, Oak Ridge National Laboratory, Apri l 24, 1961.
"
W. R . Gambill and R . D. Bundy, "High-Flux Heat Transfer Charac t e r i s t i c s of Pure Ethylene Glycol i n Axial and Swirl Flow," A.1.Ch.E. Journal, 9 ( Q : 5 5 3 9 (January 1963).
Vortex Gas Dynamics
1. J. L. Kerrebrock and R . V . Meghreblian, "An Analysis of Vortex Tubes f o r Combined Gas-Phase F iss ion Heating and Separation of t h e Fiss ion- able Material ," USAEC Report ORNL CF-57-11-3 (Revision l), Oak Ridge National Laboratory, Apri l '11, 1958.
111
2. J. L. Kerrebrock and P. G . Lafyatis, "Analytical Study of Some Aspects of Vortex Tubes for Gas-Phase Fiss ion Heating," USAEC Report ORNL CF- 58-7-4, Oak Ridge National Laboratory, Ju ly 1958.
3. J. L. Kerrebrock and J. J. Keyes, Jr., "A Preliminary Experimental Study of Vortex Tubes f o r Gas-Phase Fiss ion Heating," USAFX: Report om-2660, Oak Ridge National Laboratory, February 1959.
4. J. J. Keyes, Jr., and R. E . D i a l , "An Experimental Study of Vortex Flow fo r Application t o Gas-Phase Fiss ion Heating, Oak Ridge National Laboratory, April 1960.
USAEC Report 0R~L-2837,
5. J. J. Keyes, Jr., "An Experimental Study of Gas Dynamics i n High Velocity Vortex Flows, 'I Proceedings of the 1960 Heat Transfer and Fluid Mechanics I n s t i t u t e , pp. 31-36, Stanford University Press, Stanford, Cal i fornia .
6. J. L. Kerrebrock and R . V. Meghreblian, "Vortex Containment f o r the Gaseous Fiss ion Rocket," J. Aerospace Science, 28: 710-724 (1961).
7 . J. J. Keyes, Jr., "An Experimental Study of Flow and Separation i n Vortex Tubes with Application t o Gaseous Fiss ion Heating," Am. Rocket Soc . Journal, 1204-1210 (September 1961).
Vortex Magnetohydrodynamics
1. T . S. Chang, "Magnetohydrodynamic S t a b i l i t y of Vortex Flow - A Non- d iss ipa t ive , Incompressible Analysis ,'I USAEC Report O m TM-402, Oak Ridge National Laboratory, 'October, 23, 1962. ri
2. J. J. Keyes, Jr., "Some Applications of Magnetohydrodynamics i n Con- f ined Vortex Flows," USAM: Report ORNL TM-479, Oak Ridge National Laboratory, February 28, 1963.
Turbuleat Transport Studies
1. R. P. Wichner and F. N . Peebles, "Determination of the Six Reynolds Stresses by the Llnearized-Response Hot -Wire Anemometer, If pp. 361-370 i n Developments i n Theoretical and Applied Mechanics, Vol. 1, Plenum Press, New York, 1963.
Boundary -Layer Trans i e n t Phenomena
1. J. J. Keyes, Jr., and J. E. Mott, "An Invest igat ion of Thermal- Transient Decay at, a Fluid-Solid Interface and i n Turbulent Flow Through Circular DUcts9" USAEC Report ORNL-2603, Oak Ridge National Laboratory, December 2, 1958.
2. R . A . Suehrstedt e t a l . , "Applications of Concepts of Penetration Theory t o Heat Transfer," USAEC Report KT-378, MIT Practice School, Oak Ridge Gaseous Diffusion Plant , December 1958.
112
3. A. E. Higinbotham e t a l . , "Study of Unsteady Heat Transfer Between a Surface and a Flowing Fluid," USAM: Report KT-397, MIT Pract ice School, Oak Ridge Gaseous Diff'usion Plant , May 1959.
4. J. J. Keyes, Jr., and A. I. Krakoviak, "High-Frequency Surface Thermal Fatigue Cycling of Inconel a t 1405'F," Nucl. Sei . Eng., 9: 462474 (1961).
Progress Reports
1. H. W. Hoffhan e t al., Fundamental Studies i n Heat Transfer and Fluid Mechanics, S ta tus Report July 1, 1959 - February 29, 1 9 6 0 , ~ ~ USAM: Report ORNL CF-60-10-6, Oak Ridge National Laboratory, October 1960.
I f
.
.
ORNL-TM- 915 INTERNAL DISTRIBUTION
1. L. G. Alexander 2. R . A. Armistead, Jr. 3. S. E . Beall 4. M. Bender 5. E . S. Bet t i s 6. E . P. B l i z a r d 7. A. L. 3och 8. E . G. Bohlmann 9. G. E. Boyd
10. C . A. Brandon 11. R. B. Briggs 12. R . D. Bundy 13. S. J. Claiborne, Jr. 14. T . S. Chang 15 . T . E . Cole 16. J. W . Cooke 17. W . B. C o t t r e l l 18. F. L. Cul ler 19. R . E . Dial 20. J. A. Edwards 21. D. M. Eissenberg 22. W. K. Ergen 23. A. P. Fraas 24. J. H. Frye, Jr. 25. U. C . Fulmer, Jr. 26. W. R . Gambill 27. W. F. Gauster 28. B. L. Greenstreet 29. W . R . Grimes 30. A. G. Grindell 31. R. P. Hammond 32. P. N. Haubenreich 33. P. H. Hayes 34. J. W . H i l l
35-44. H. W . Hoffman 45. W. H. Jordan
46-50. J. J. Keyes, Jr. '
51 . G. J. Kidd, Jr. 52. R . B. Korsmeyer 53. A. I. Krakoviak 54. K. A. Kraus
55 * 56. 57. 58. 59. 60. 61. 62. 63 * 64. 65 * 66. 67 68. 69 70 71 * 72 73 * 74 75 * 76. 77 78. 79 80. 81. 82. 83 84. 85. 86. 87. 88.
89-91. 92-93
94-96. 97 *
98
J. A. Lane G . E . Larson M. E . Laverne C . G. Lawson J. Lones M. I. Lundin R . N . Lyon H. G. MacPherson R . E . MacPherson H. C . McCurdy W. B. McDonald A. J. Mil ler R . L. Mil ler T . C . Min F. N. Peebles . A. M. Perry M. W . Rosenthal A. F. Rupp G. Samuels H. W . Savage A. W . Savolainen M. J. Skinner A. H. Snel l I. Spiewak R. M. Summers B. J. Sutton J. A. Swartout W. C . Thurber D. B. Trauger D. H. Wallace J. S. Watson A. M . We inbe rg G. D. Whitman Gale Young Central Research Library (CRL) Y -12 Document Ref e renc e Sect ion ( DRS) Laboratory Records Department (LRD) Laboratory Records Department - Record Copy (LRD-RC) Reactor Division Library, Bldg. 9204-1
EXTERNAL DISTRIBUTION
a,
0
99-113. Division of Technical Information Extension (DTIE) 114. Research and Development Division, OR0
115-116. Reactor Division, OR0
,i' i
117. 118. 119. 120. 121. 122.
124. 123.
125. 126. 127. 128.
129. 130. 131. 132. 133 - 136. 137
138-139. 140-141.
134-135.
142. 143. 144. 145. 146.
147-149.
150-151.
152.
153. 154. 155 * 156.
157-158.
159.
160.
161.
162.
163.
' i k. I -
A.E.G., AEG Hochhaus, Frankfurt , Germany (M. Schul le r ) Aero jet-General Nucleonics (K. Johnson) Allis-Chalmers Manufacturing Co. (K. F. Neusen) Alsthom, Par i s , France (M. P. Domenjoud) Argonne National Laboratory (Dr. Paul Lot tes ) Argonne National Laboratory ( D r . M. Pe t r i ck ) At lan t ic Research Nuclear C o g . ( D r . M. Markels, Jr .) Atomics In te rna t iona l ( D r . L. Bernath) Babcock & Wilcox Co. (D. F. Judd) Brookhaven National Laboratory (Dr. 0. E . m e r ) Centre d'Etudes Nucl&aires, Grenoble, France (M. Mondin) Centre d'Etudes NuclAaires de Saclay, G i f sur Yvette, France (M. J. Horowitz)
CHEN, Roma, I t a l y (Prof. F. Ippo l i to ) #
CISE, Milano, I t a l y (Prof. M. S i l v e s t r i ) Compagnie Francaise, Gagneux, France (M. Lefranc) Dartmouth College, Thayer School of Engineering ( D r . G. B. Wallis) ?
Dynatech Corporation ( D r . M. Suo) EURATOM, Bruxelles, Belgium ( D r . P. Kruys) EURATOM, Ispra , I t a l y (M. R . Morin) FIAT, Torino, I t a l y (M. G. Cesoni) General E lec t r i c Co., Atomic Power Equipment Dept. ( D r . S. Levy) General E l e c t r i c Co., Advanced Technology Laboratory ( D r . Novak Zuber ) Geoscience Ltd. ( D r . H. F. Poppendiek) Hanford Atomic Products Operation (Dr. John Batch) MAN, Ntirnberg, Germany ( D r . Mayinger) Massachusetts I n s t i t u t e of Technology ( D r . P. G r i f f i t h ) Massachusetts I n s t i t u t e of Technology (Dr. W . M . Rohsenow) National Aeronautics & Space Administration, Lewis Research Center (Library) National Aeronautics & Space Administration, L e w i s Research Center [R. Weltmann (SEPO)] North Carolina S ta t e University, Department of Chemical Engineering ( D r . J . K. F e r r e l l ) Reactor Centrum Nederland, s Gravenhage, Netherlands (M. Pe l se r ) SNECMA, Division Atomique, Suresnes, France (M. Four&) Technische Hogeschool, Eindhoven, Netherlands (Prof. Bogaardt) University of Minnesota, Department of Chemical Engineering ( D r . H. S. I sb in )
U. S. Atomic Energy Commission, Division of Reactor Development (R. M. Scroggins) U. S. Atomic Energy Commission, Division of Reactor Development (Office of C iv i l i an Power, Water Reactors Branch)
U. S. Atomic Energy Commission, Division of Reactor Development (Office of Army Reactors, Technical Evaluation Branch)
U. S. Atomic Energy Commission, Division of Reactor Development (Office of Nuclear Safety, Research and Development Branch)
U. S. Atomic Energy Commission, Division of Safety Standards (R. Irapara) Westinghouse E lec t r i c Corporation, Atomic Power Division ( D r . L. S. Tong)